banatvala rubella viruses

Rubella Viruses

PERSPECTIVES IN MEDICAL VIROLOGY

Volume 15

Series Editors

A.J. Zuckerman

Royal Free and University College Medical SchoolUniversity College London

London, UK

I.K. Mushahwar

Abbott LaboratoriesViral Discovery GroupAbbott Park, IL USA

Rubella Viruses

Editors

Jangu Banatvala

King’s College London Medical School at Guy’sKing’s College and St Thomas’ Hospitals

London, UK

Catherine Peckham

Centre for Paediatric Epidemiology and BiostatisticsUCL Institute of Child Health

London, UK

Amsterdam – Boston – Heidelberg – London – New York – Oxford – ParisSan Diego – San Francisco – Singapore – Sydney – Tokyo

Elsevier

Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK

First edition 2007

Copyright r 2007 Elsevier B.V. All rights reserved

No part of this publication may be reproduced, stored in a retrieval system

or transmitted in any form or by any means electronic, mechanical, photocopying,

recording or otherwise without the prior written permission of the publisher

Permissions may be sought directly from Elsevier’s Science & Technology Rights

Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333;

email: [email protected]. Alternatively you can submit your request online by

visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting

Obtaining permission to use Elsevier material

Notice

No responsibility is assumed by the publisher for any injury and/or damage to persons

or property as a matter of products liability, negligence or otherwise, or from any use

or operation of any methods, products, instructions or ideas contained in the material

herein. Because of rapid advances in the medical sciences, in particular, independent

verification of diagnoses and drug dosages should be made

Library of Congress Cataloguing-in-Publication Data

A catalog record for this book is available from the Library of Congress

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN-13: 978-0-444-50634-4

ISBN-10: 0-444-50634-9

ISSN: 0168-7069

Printed and bound in The Netherlands

07 08 09 10 11 10 9 8 7 6 5 4 3 2 1

For information on all Elsevier publications

visit our website at books.elsevier.com

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

Historical Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Molecular Virology of Rubella VirusM.-H. Chen and J. Icenogle . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Clinical Features: Post-Natally Acquired RubellaJ.E. Banatvala . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Laboratory Diagnosis of Rubella and Congenital RubellaJ.M. Best and G. Enders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Rubella VaccineS. Reef and S.A. Plotkin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

Rubella Epidemiology: Surveillance to Monitor and Evaluate CongenitalRubella Prevention StrategiesC. Peckham, P. Tookey and P. Hardelid . . . . . . . . . . . . . . . . . . . . 95

Future Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

List of Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

v

This page intentionally left blank

Preface

Since its introduction in the 1960s, rubella vaccination has substantially changedthe epidemiology of rubella and congenital rubella nationally and internationally.Worldwide, hundreds of thousands of congenital rubella births have been averted,but the infection continues to inflict a considerable burden in many countries.Different national and regional approaches to rubella control, the ease with whichinfection can be imported across national boundaries, and the fact that infection isusually mild and frequently asymptomatic, mean that the elimination of rubella willbe no easy task. Success in controlling rubella will depend on a sustained politicaland economic commitment to vaccination and surveillance over the long term. Thisbook highlights the enormous challenges to be faced, through an up-to-date ap-praisal of the molecular, clinical, laboratory and epidemiological features of rubellaand congenital rubella in the vaccine era.

Jangu BanatvalaLondon, UK

Catherine S. PeckhamCentre for Paediatric Epidemiology and Biostatistics

UCL Institute of Child Health

London, UK

vii

Historical Introduction

Rubella was originally known by its German name ‘‘Roteln’’ having been describedoriginally by two German physicians. For many years German measles was con-fused with other infections causing a rash, particularly measles and scarlet feverand even today a diagnosis on clinical grounds alone is notoriously inaccurate (seeChapter 2). German measles was, in due course, recognised as a distinct disease byan International Congress in Medicine in London in 1881 and the infection des-ignated as ‘‘rubella’’ was accepted at about that time (reviewed by Best and Ban-atvala, 2004).

It was not until 1941 that an Australian ophthalmologist, Norman McAlisterGregg in his epoch-making paper produced evidence, which showed that if ac-quired in early pregnancy, rubella caused developmental defects in the fetus (Gregg,1941). In 1941, there were extensive outbreaks of rubella among troops in trainingin New South Wales, Victoria and Queensland, being mobilised during the SecondWorld War. Infection spread to the civilian population and it is possible that troops

ix

may have infected their wives on returning home prior to serving overseas. Greggnoted the presence of cataracts, usually bilateral, microphthalmia and the ‘‘salt andpepper’’ appearance of a characteristic retinopathy occurring in congenitally in-fected infants. A high percentage of infants also had cardiac anomalies, failed tothrive and in due course many were found to have severe perceptive deafness.

Gregg was the first to recognise that a virus infection in humans could result inembryopathy and that damage was the result of infection acquired in early fetal life.Although Gregg’s findings were not universally accepted initially, other retrospec-tive studies, not only in Australia but in other countries, subsequently confirmedhis findings (reviewed by Hanshaw et al., 1985). The earlier studies were retro-spective and, as might be expected, were at risk of overestimating the incidence ofdefects, since the starting point for investigations was the delivery of infants withcongenital anomalies, whose mothers reported a rubella-like rash during preg-nancy. However, prospective studies carried out in the 1950s and early 1960s, inwhich the starting point was a mother with a rubella-like illness, rather than aninfant with congenital anomalies, revealed a much lower incidence of congenitalmalformation, varying from 10% to 54% (Hanshaw et al., 1985). But such studiesunderestimated the incidence of rubella-induced congenital anomalies, since theydid not then have the benefit of laboratory confirmation of the clinical diagnosis.Undoubtedly, many women with rash in pregnancy were included and were de-livered of healthy babies, but had infections other than rubella. However, subse-quent studies showed that following virologically confirmed rubella in the firsttrimester, the fetus is almost invariably infected, and about 80–85% of infants aredamaged (see Chapter 2). Thus, despite the limitations of the earlier retrospectivestudies, the original observations were a reasonably accurate assessment of the riskto the fetus following maternal rubella, but this was due to the fact that the ex-tensive Australian epidemics in the 1940s were caused by rubella rather than byother infections-inducing rashes.

Rubella virus was isolated for the first time in 1962 by two groups workingindependently in the USA employing different cell culture techniques (Parkman etal., 1962; Weller and Neva, 1962). This discovery led to the development of se-rological tests, which could be used to determine rubella virus immunity, andconfirm or refute the clinical diagnosis of rubella among women who were exposedto, or who developed rubella during pregnancy. Laboratory investigations werenow also available to diagnose, and be used during follow-up studies on infantswho had been exposed to rubella in utero.

In 1963/1964, the USA experienced one of the most extensive recorded out-breaks of rubella, which led to a greater understanding of the pathogenesis ofcongenitally acquired rubella as well as a fuller appreciation of the clinical featuresand sequelae. It was estimated that somewhere of the order of 20,000–30,000 ru-bella damaged babies resulted from this epidemic and that these infants experienceda generalised and persistent infection, occurring not only in utero but extending intoinfancy (see Chapters 2 and 3). Although it was thought that the wider spectrum ofanomalies noted among infants with congenitally acquired rubella following the

Historical Introductionx

1963/1964 epidemic, termed at that time ‘‘the expanded rubella syndrome’’, was anew phenomenon, careful inspection of clinical notes of infants with congenitallyacquired disease from previous but smaller outbreaks showed that many of the so-called new clinical features had been observed previously.

The extensive nature of the 1963/1964 rubella epidemic emphasised the impor-tance of attempting to prevent infection by the development of vaccines, and during1965/1966 attenuated vaccines were developed and the first vaccine trials started.

During the next few years, different vaccines were licensed and rubella vac-cination programmes commenced in the USA, UK and subsequently in other partsof the world (see Chapter 4) making congenitally acquired rubella a preventabledisease. Current vaccination programmes have reduced the incidence of rubellamarkedly in many industrialised countries and WHO has now put forward pro-grammes to eliminate rubella, not only in industrialised countries, but also in manydeveloping countries (see Chapters 4 and 5).

Table 1 illustrates some of the principal developments in the history of rubella.

Table 1

Main developments in history of rubella

1881 International Congress on Medicine recognised rubella as a

distinct disease

1941 Gregg in Australia recognises teratogenic effects

1962 Rubella virus isolated in cell culture. Neutralisation tests

developed

1963–1964 Extensive European and USA epidemics. 12.5 million rubella

cases, 11,000 fetal deaths and 20,000 CRS cases in USA

1969 and 1970 Attenuated rubella vaccines licensed in USA and UK (USA

universal childhood programme; UK selective vaccination of

prepubertal school girls)

1971 MMR licensed in USA

1978, 1979 and 1983 Severe UK rubella epidemics mostly involving adolescent and

young adult males but some pregnant women

1986 Rubella virus genome sequenced

1988 UK policy augmented by offering MMR to pre-school

children of both sexes

1989 USA introduced a two-dose vaccination at age 12–15 months

and at age 4–5 years or 11–12 years

1989–1991 Resurgence of rubella in USA

1996 In UK, schoolgirl vaccination discontinued but second dose

of MMR introduced for children aged 4–5 years

2000 WHO recommends immunisation policies for elimination of

CRS

2002 123 (57%) of 212 of countries and territories include rubella

vaccination in national immunisation programmes

Source: Modified from Lancet 2004; 363: 1127–1137.

Historical Introduction xi

References

Best JM, Banatvala JE, Rubella. In: Principles and Practice of Clinical Virology (Zuckerman

AJ, Banatvala JE, Pattison JR, Griffiths PD, Schoub BD, editors) Chapter 12, Chichester:

Published by John Wiley & sons; 2004; pp. 427–457.

Gregg NM. Congenital cataract following German measles in mother. Trans. Ophthalmol

Soc Aust 1941; 3: 35–46.

Hanshaw JB, Dudgeon JA, Marshall WC. Viral Diseases of the Fetus and Newborn. 2nd ed.

Philadelphia: W.B. Saunders; 1985.

Parkman PD, Buescher EL, Artenstein MS. Recovery of rubella virus from army recruits.

Proc Soc Exp Biol Med 1962; 111: 225–230.

Weller TH, Neva A. Propagation in tissue culture of cytopathic agents from patients with

rubella-like illness. Proc Soc Exp Biol Med 1962; 111: 215–225.

Historical Introductionxii

Rubella Viruses

Jangu Banatvala and Catherine Peckham (Editors)

r 2007 Elsevier B.V. All rights reserved

DOI 10.1016/S0168-7069(06)15001-6

1

Chapter 1

Molecular Virology of Rubella Virus

Min-Hsin Chen, Joseph IcenogleDivision of Viral Diseases, Centers for Disease Control and Prevention, 1600 CliftonRoad, Atlanta, GA 30333, USA

Introduction

Rubella virus is the sole member of the Rubivirus genus in the Togaviridae family,which also includes the Alphavirus genus (Chantler et al., 2001). Alphaviruses andrubella virus share a similar genetic organization and replication strategy. Alpha-viruses are transmitted to humans by arthropods while rubella virus is transmittedbetween humans; humans are the only known hosts for rubella virus. An incom-plete list of alphaviruses and the human diseases they are known to cause is RossRiver virus (fever, arthritis, rash), Semliki Forest virus (fever, encephalitis),Venezuelan Equine Encephalitis virus (fever, encephalitis), and Sindbis virus (fever,arthritis, rash) (Griffin, 2001). There is little sequence homology between rubellavirus and any alphavirus, limiting the possibilities for studying the relationshipsbetween rubella virus and alphaviruses by simple phylogenetic analysis. Sincerubella virus is more difficult to culture than some alphaviruses (e.g. Sindbis virus),equivalent information for alphaviruses and rubella virus is often lacking. Thus, inthe present chapter, some information is presented from work done with alpha-viruses and then extrapolations are made to rubella virus.

Rubella virus is a potent, infectious, teratogenic agent, which continues to causedevastating epidemics of congenital rubella syndrome (CRS) in much of the world(Cooper, 1985; Lee and Bowden, 2000). The virus was isolated in 1962 and by 1969an attenuated, live vaccine was licensed (Parkman et al., 1962; Weller and Neva,1962). Viruses isolated from CRS cases are identical to viruses from postnatalrubella cases (Lee and Bowden, 2000). In utero infection with rubella virus in thefirst trimester usually produces one or more of a set of specific pathologies in the

dx.doi.org/10.1016/S0168-7069(06)15001-6.3d

fetus. Although there have been some molecular studies in cell lines directed towardthe pathogenesis of rubella virus, the molecular details of the pathogenesis (i.e. themechanism(s) of teratogenicity of rubella virus) in humans are poorly understood.

Structure of the virion

Rubella virus virions are particles about 70 nm in diameter with a lipid envelopecontaining two viral glycoproteins, E1 and E2, and a nucleocapsid, containing thepositive-strand RNA molecule and the capsid protein, C (Dominguez et al., 1990;Risco et al., 2003; Zheng et al., 2003c) (Fig. 1A). The structure of the rubella virionhas not been precisely determined by fitting crystallographic structures of the in-dividual proteins into cryoelectron microscopic maps, as has been done for alpha-virus virions (Zhang et al., 2002; Gibbons et al., 2004). These studies of the virionstructures of various alphaviruses have shown interactions between the C proteinand RNA, the E1–E2 glycoprotein heterodimers organized into trimers on thevirion surface, and the overall structure of the virion. Major conformationalchanges in the E1–E2 heterodimer occur at low pH and upon interactions withcells, exposing the fusion protein on the E1 protein of Sindbis virus (see Kuhn et al.,2002 for discussion; Paredes et al., 2004). These studies likely provide an approx-imate model for the structure of the rubella virion and conformational changes thatlikely occur in the rubella virion (Fig. 2). One should also note that there aresignificant similarities in structure and function between the large envelope proteinof alphaviruses and the envelope protein of flaviviruses, which are more distantlyrelated to the alphaviruses than is rubella virus (Modis et al., 2004).

The C proteins in the rubella virion exist as disulfide-linked homodimers,(although the dimerization is not required for formation of viral particles (Leeet al., 1996)). Analysis of the amino acid sequence of the rubella virus C proteinsuggests that the N-terminal half of this protein interacts with RNA, because it ishydrophilic and rich in prolines and arginines (reviewed in Frey, 1994). The majorRNA-binding domain in the C protein has been located within amino acid residues28–56, but other regions, including the C-terminus, might also be involved in en-hancing the interaction (Liu et al., 1996). The C-terminus of the C protein is veryhydrophobic and retains the putative signal peptide of the E2 protein (reviewed inFrey, 1994). Therefore, the C proteins in virions are anchored to the viral mem-brane by their C-termini, with the N-terminal regions inside the viral envelope andlikely contacting the RNA. Since the C proteins are bound to the viral membrane,nucleocapsid assembly/disassembly for rubella virus may well occur by pathwaysdifferent from those of alphaviruses (see below). It is possible that the N-terminalregion of the C protein interacts with the cytoplasmic tail of E1 and is involved inbudding (Hobman et al., 1994).

The E1 and E2 glycoproteins exist as heterodimers in the rubella virion. The E1protein is a class 1 transmembrane protein with three N-linked glycosylation sites inthe N-terminal half of the protein. Although glycosylation of E1 does influence the

M.-H. Chen, J. Icenogle2

Fig. 1 Electron micrographs from rubella virus infected Vero cell culture. Panel A. Electron micrograph

of rubella virus particles in infected Vero cell culture (isolate JV 13R-23) showing dense core and

surrounding lipid bilayer (one virion is indicated by arrow). The virus was isolated from human case of

rubella. (Electron micrograph by Dr. Elena I. Ryabchikova, Dr. G. I. Tiunnikov, and Dr. Vladimir S.

Petrov (State Research Center of Virology and Biotechnology VECTOR, Russia) with permission). Bar

corresponds to 60 nm. Panel B. Rubella virus replication complex in Vero cells. Each complex is a

modified lysosome-containing vesicles (arrow heads) that line the inner membrane of a cytopathic

vacuole (v). The rough endoplasmic reticulum is indicated by open arrows. The bar represents 200 nm.

Figure is from Lee et al. (1999), with permission. The closed arrows with tails indicate rubella virus core

particles, which were seen in association with the periphery of cytopathic vacuoles in this study, but this

observation was not found in another study (Risco et al., 2003).

formation of infectious virus, likely because glycosylation is required for properfolding of the E1 protein, glycosylation does not play a role in the antigenicity ofthe virion (Qiu et al., 1992; Ramanujam et al., 2001). The E1 protein containsimportant functional domains. The fusion peptide, which is exposed during fusionof the virion membrane with cellular membranes during entry into cells, is likelylocated at amino acid 81–109 of the E1 protein (Qui et al., 2000). Amino acidresidues 81–109 of E1 are also likely important for the interaction between E1 andE2 (Yang et al., 1998). The E1 protein is important in the antigenicity of the rubellavirus virion; it contains antigenic sites, as defined by monoclonal antibody binding,in the region between amino acids 245 and 284. In patients who have had rubella, ahemagglutination inhibition and neutralization epitope maps to amino acids208–239 (Ho-Terry et al., 1984; Chaye et al., 1992; Wolinsky et al., 1993; Cordobaet al., 2000). The E2 protein is also a class 1 transmembrane protein, which isheavily glycosylated, both N- and O-linked (Nakhasi et al., 2001). The E2 protein

Fig. 2 Proposed configuration of rubella E1 and E2 proteins on the surface of rubella virus virions by

analogy with the alphaviruses. The configuration of the E1 and E2 proteins on the surface of alphavirus

virions and the position of the fusion peptide (FP) on the E1 protein at neutral pH, and the proposed

configuration at acid pH are shown (e.g. dengue virus). Although no high-resolution structures have

been determined for rubella virus virions, since rubella virus virions have a similar composition to that of

alphaviruses, it is likely that rubella E1 and E2 proteins will have a similar structure and will undergo

similar transitions to those shown. Figure adapted from Kuhn et al. (2002), with permission. Proposed

configurations of rubella proteins are the opinion of the authors of this chapter.


remains linked to the signal peptide of the E1 protein (Hobman et al., 1997).Although the E2 protein of Sindbis is known to be involved in receptor binding,receptor binding has not been mapped to the rubella virus E2 protein (Myles et al.,2003; discussed in Paredes et al., 2004).

There is significant sequence variation among the currently circulating rubellaviruses; although a specific antigenic site in the E2 protein recognized a monoclonalantibody in western blots is not conserved, overall there is only one serotype ofrubella viruses (Zheng et al., 2003c). There is evidence that the E1 protein alonemay be useful as a vaccine (Perrenoud et al., 2004). All evidence indicates thatprotection from all circulating viruses is achieved by vaccination using the current,most common, live-attenuated vaccine strain of rubella virus, RA27/3.

Genome organization

The rubella virus genome is a single-stranded RNA of 9762 nucleotides (nts) inlength (Dominguez et al., 1990), which is 50 capped and 30 polyadenylated andserves as a messenger RNA (mRNA) during infection. One of the significantcharacteristics of the rubella virus genome is its high G+C content (about 70%),which is the highest of any RNA virus thus far sequenced. The genome containstwo open reading frames (ORFs): the 50 proximal ORF encoding nonstructuralproteins (NSPs), including a RNA-dependent RNA polymerase that is essential forviral genome replication, and the 30 proximal ORF encoding the viral structuralproteins (SP) (the capsid protein and the two envelope glycoproteins) (Fig. 3A).The untranslated regions (UTRs) in the rubella virus genome include a 40-ntsequence at the 50 end (50 UTR), a�118-nt sequence in the intragenic region betweenthe ORFs and a 59-nt sequence at the 30 end (30 UTR) (reviewed by Frey, 1994).

Attachment and entry

The effects of rubella virus replication on the host cell are dependent on the type ofcell infected (Risco et al., 2003). Thus, one must be careful not to extrapolate fromone cell line to another and, more importantly, from replication of rubella virus incell lines to replication of rubella virus in humans. Nevertheless, enough is knownabout the attachment, entry, and replication of rubella virus and related viruses incell lines to provide a reasonable overall understanding of rubella virus attachment,entry, and replication.

Attachment of rubella virus to cellular receptors presumably occurs via bindingsites on the E2 and/or E1 glycoprotein molecules, even though the cellular receptorfor rubella virus has not been identified nor has the receptor-binding site on one orboth of the glycoproteins. Both cellular receptors and receptor-binding sites on theglycoproteins have been identified for various alphaviruses (discussed in Kuhnet al., 2002; Myles et al., 2003). Drugs that inhibit specific endocytic pathways have

Molecular Virology of Rubella Virus 5

5’(+)

P200P200

(-) 3’

P200P200

Host protein(s)?

UUU(+) Genomic RNA

Subgenomic RNA

Translation of 5’ ORF

Genome replication

Genome replication and Transcription

(+) AAA

(+) AAA

NSP AAA 3’SP

P90

P15

0

P90

P15

0

UUU 5’P200P200

NSP cleavage

Translation 3’ ORF

(-) Genomic RNA

Nonstructural Protein Structural Protein

Cp150 p90

Methyl transferase

Protease

Helicase

RNA binding domain

Signal peptide (and transmembrane domain)

Transmembrane domain

Fusion peptide

AAA 3’5’

(A)

(B)

RNA-dependent RNA polymerase

E1E2

Fig. 3 Rubella Virus Genome Organization (A) and RNA replication (B). (A) The two open reading

frames (ORFs) are denoted by open boxes and the three untranslated regions (UTRs) in the genome are

indicated by solid lines. The 50 cap structure is indicated by a solid circle and 30 polyadenylation is

indicated by AAA. Individual proteins, after processing of the polyproteins from the 50 and 30 ORFs are

shown below the genome. Known functional domains in each protein are denoted by boxes. (B) Rubella

genome replication. The minus- and plus-strand RNAs are indicated by (�) and (+), respectively. The

putative promoters for each RNA species on the plus- or minus-strand genomic RNA are denoted by

solid boxes. Translation of 50 and 30 ORFs is indicated by the gray arrow and the progress of RNA

synthesis by rubella replication proteins is indicated by arrows with dash lines.


been used to investigate rubella virus entry into Vero cells with the result that themajor pathway utilized by rubella virus has been identified as clathrin-mediatedendocytosis (also see Lee and Bowden, 2000 and Kee et al., 2004). It is reasonableto assert that a conformational change(s) in the E1 and/or E2 proteins occursfollowing virus attachment which exposes the fusion peptide in E1, allows thefusion of viral and cellular membrane, and allows the entry of the RNA (or thenucleocapsid) into the cytoplasm (Fig. 2). Since the nucleocapsid is anchored to theviral membrane via the C-termini of its C proteins, it is not clear whether mem-brane fusion and nucleocapsid disassembly are separate events. In the case ofalphaviruses these conformational changes are well understood and occur at lowpH (see Kuhn et al., 2002 for discussion).

Intracellular site of RNA replication

Early in infection, rubella virus infected cells contain cytopathic vacuoles whichthemselves contain RNA-replication complexes (Magliano et al., 1998) (Fig. 1B).The number and size of rubella virus-induced cytopathic vacuoles depend on thetype of cell infected (Risco et al., 2003). Reports of colocalization of the C proteinwith cytopathic vacuoles during RNA replication and the enhancement of poorlyreplicating, rubella virus self-replicating replicons by the C protein suggestinvolvement of the C protein in RNA replication (Lee et al., 1994; Chen andIcenogle, 2004). Self-replicating replicons are modified rubella virus RNAs whichhave a partial genome; usually replicons have the SP ORF replaced by another gene.

Transcription and translation

The genomic RNA of rubella virus can be an mRNA for translation of the NSPORF, a template for the minus-strand genomic RNA synthesis, or encapsidated byviral SP to form nucleocapsids during assembly. Only the NSP ORF is translatedfrom genomic RNA; a polyprotein of molecular weight �200 kDa is produced. TheNSP ORF encodes the major viral proteins responsible for viral genome replication(i.e. synthesizing plus- and minus-strand genomic RNA) and synthesizing RNAsfor translation (viral plus-strand genomic RNA and a subgenomic RNA).Production of viral proteins is dependent on the cellular translational machineryand cis-acting elements in the mRNAs are involved in regulation of translationefficiency. The 50 stem-loop structure, for example, is important for the translationefficiency of the NSP ORF (see Pogue et al., 1993; Pugachev and Frey, 1998a). Theeffect of a 30 terminal stem-loop on the translation efficiency is unclear. Enhancedtranslation of a monocistronic minireplicon by the 30 stem-loop was observed(Pogue et al., 1993), but no significant effect on viral NSP translation or fromsubgenomic RNA was seen when using a self-replicating rubella virus replicon(see below; Chen et al., 2004; Pappas et al., 2006). The rubella virus subgenomic


RNA encodes the 30 proximal ORF (the SP ORF). The subgenomic RNA onlyserves as an mRNA for the production of viral SP. The 50 UTR (�76 nts inF-therien) of the subgenomic RNA was found to be necessary for optimal trans-lation efficiency in replicons (Pappas et al., 2006). Although a second in-frame startcodon with a consensus Kozak context exists in the SP ORF (located at the 9thamino acid downstream from the start codon of the first ORF), translation of SPORF is preferentially initiated from the first AUG (authors’ unpublished data). InSindbis virus, a hairpin structure located downstream of the AUG start codonwhich initiates translation for the C protein, was important for enhancing trans-lation of subgenomic RNA (Frolov and Schlesinger, 1996). A similar hairpinstructure in rubella virus C protein coding region was found. Unlike Sindbis virus,this structure attenuates translation from SG RNA (Pappas et al., 2006).

Both genomic and subgenomic RNAs are transcribed by viral replication pro-teins using the minus-strand genomic RNA as a template. In infected cells, the plus-strand RNA species are present in higher molar ratio than the minus-strand RNA,and more subgenomic RNA is synthesized than plus-strand genomic RNA (molarratio of 1.6:1) (Hemphill et al., 1988). Interestingly, a study of the 30 cis-actingelements using a self-replicating rubella replicon encoding an antibiotic resistantgene showed that the amount of plus-strand RNA correlated better with thenumber of cells surviving antibiotics than did the amount of minus-strand RNA,suggesting the synthesis of plus-strand RNA, or transcription, may be the key stepof a successful viral replication (Chen et al., 2004).

Genome replication

Replication of the rubella virus genome begins with the synthesis of the NSPpolyprotein. This polyprotein is presumed to recognize a promoter at the 30 ter-minus of the genome and synthesize a minus-strand genomic RNA, which is pro-posed to form a double-stranded replicative intermediate with the genomic RNA.The minus-strand genomic RNA is then used as a template for the synthesis of boththe genomic RNA and the subgenomic RNA. The subgenomic RNA is synthesizedby utilizing an internal promoter (subgenomic promoter) and is colinear with the 30

terminal 3326 nts of the genomic RNA (reviewed by Frey, 1994) (Fig. 3B).

Regulatory elements in RNA

Comparison of nucleotide sequences of alphaviruses revealed four conserved re-gions which might be involved in virus replication by functioning as cis-actingregulatory signals (reviewed by Strauss and Strauss, 1986). These include the 50

terminus, a 51-nt domain near the 50 end, the 21-nt sequence found in the ‘‘junctionregion’’ upstream from the subgenomic RNA start site and the 19-nt sequenceadjacent to the 30 poly(A) tract. Stretches of nucleotides sharing homology with thefirst three of these four conserved regions are present in the rubella virus genome


(Dominguez et al., 1990), which suggests the importance of these regions in To-gavirus genome replication. Development of a rubella infectious cDNA clone(Wang et al., 1994; Pugachev et al., 1997) and self-replicating replicons containingreporter genes (Tzeng and Frey, 2003; Chen et al., 2004) allows the study of thebiological activities of these regulatory elements.

Similar to Sindbis virus, the 50 end of the genomic RNA of rubella virus con-tains a 14-nucleotide single-stranded leader followed by a stem-and-loop structure(nt 15–65) with a potential pseudoknot structure (reviewed by Frey, 1994). Amutagenic study of this element using a rubella virus infectious clone showed thatonly the AA dinucleotide at nucleotide 2 and 3 was essential for the viability of thevirus; the stem-loop structure was important for the level of production of the NSPproteins but maintenance of the complementary sequences of this structure was notrequired for plus-strand RNA synthesis (Pugachev and Frey, 1998a).

While mutagenesis in most of the 50 UTR was tolerable, most of the 30 UTRwas required for viral viability except the 30 terminal 5 nts and poly(A) tail (Chenand Frey, 1999; Chen et al., 2004). The RNA elements in the E1 coding region wererequired for efficient replication. For example, the 30 terminal 305 nts were nec-essary for optimal replication. Maintenance of a GC-rich stem-loop structure at theC-terminus of E1 coding region, which was previously shown to interact with threehost factors, was not necessary for virus viability.

The RNA complementary to the �118-nt intragenic region between 50 and 30

ORFs is presumed to function as a subgenomic RNA promoter. Using a modifiedrubella virus infectious RNA (Pugachev et al., 2000) and a replicon (Tzeng et al.,2001), the minimal subgenomic RNA promoter was mapped from nt �26 throughthe subgenomic RNA start site and appears to extend to at least nt +6 (transcriptionof subgenomic RNA starts at nt +1), although a larger region is required for thegeneration of virus with a wild-type phenotype. Interestingly, while the positioning ofthe rubella virus subgenomic promoter immediately adjacent to the subgenomicRNA start site is similar to that of alphavirus, it does not include the 21-nt sequencehomologous with the alphavirus subgenomic RNA promoter which is located be-tween nt �48 and �23 in the rubella virus genome (Tzeng and Frey, 2002).

Regulatory elements of RNA replication—the roles of nonstructural and structuralproteins

The NSP polyprotein, p200, is processed into p150 and p90 by a virally encodedpapain-like thiol protease (Marr et al., 1994) (Fig. 3A). The order of thesepolypeptides is NH2–p150-p90–COOH (Forng and Frey, 1995). Both methyltransferase and protease domains are embedded in p150 while the p90 includes thehelicase and RNA-dependent RNA polymerase (reviewed by Frey, 1994). BothCys-1151 and His-1272 are necessary for the catalytic activity of rubella proteasein p150 (Marr et al., 1994; Chen et al., 1996) and the cleavage of p200 betweenGly-1300 and Gly-1301 occurs both in trans and cis (Chen et al., 1996; Liang et al.,2000).


The processing of p200 is crucial for RNA replication (Liang and Gillam,2000). Mutations abolishing cleavage of p200 resulted in accumulation of minus-strand RNA but no plus-strand RNA synthesis. This indicated that uncleaved p200could function in minus-strand RNA synthesis, whereas the cleavage products p150and p90 are required for efficient plusstrand RNA synthesis and suggested thatdifferent replication proteins may recognize different promoters. A similar mech-anism has been suggested for Sindbis virus (Lemm et al., 1994). The synthesis ofminus-strand RNA by p200 is cis-preferential (Liang and Gillam, 2001), althoughrescuing replication-defective mutants by trans-complementation of the p200cleavage products can occur (Wang et al., 2002).

The p90 protein contains a conserved putative RNA-dependent RNA polym-erase domain (Kamer and Argos, 1984). There is no direct evidence confirmingwhether rubella p90 functions as RNA-dependent RNA polymerase although mu-tation of the conserved GDD tripeptide, which is characteristic of RNA-dependentRNA polymerases, in p90 affected virus viability (Wang and Gillam, 2001).Bacterially expressed rubella recombinant protein containing the C-terminus ofp150 and N-terminus of p90 (amino acid residues 1225 and 1664 of p200) wasproven to have an RNA-dependent nucleoside triphosphatase activity, suggestingthat the N-terminus of p90 could function as an RNA helicase required forunwinding RNA structures during replication (Gros and Wengler, 1996). Aretinoblastoma protein-binding motif LXCXE, was also found in, which is alsorequired for efficient virus replication (Forng and Atreya, 1999).

Interestingly, in addition to NSP, the SP of rubella virus was also found to beinvolved in RNA replication (Tzeng and Frey, 2003; Chen and Icenogle, 2004). Theminimal domain in SP required to affect replicon replication was mapped to the Cprotein. The enhanced genome replication may involve both trans- and cis-actingelements since either including the C encoding sequences in rubella virus replicons orexpressing C proteins in trans enhances replicon replication. In one study, it wassuggested that the C proteins specifically complemented rubella virus NSP, p150 (Tzengand Frey, 2003). In another study, replication of mutants with deleterious deletions inthe 30 UTRwas enhanced by co-expression of C protein, suggesting that C protein mayenhance the RNA replication at the early stage in genome replication (Chen andIcenogle, 2004). Less subgenomic RNA was produced by viruses with mutations in theproposed RNA-binding site of the C protein and resulted in poor interaction with amitochondria matrix protein p32, suggesting that C protein is also important in reg-ulating the amount of subgenomic RNA synthesis (Beatch et al., 2005).

Interactions between rubella virus proteins and cellular proteins involved in cell cycle

regulation

Rubella virus induced cytoplasmic effect and subsequent cell death have beenassociated with capase-3-dependent programmed cell death and the NSP wereassociated with rubella virus induced cell death (Pugachev et al., 1997; Pugachev


and Frey, 1998b). In addition, p90 has been shown to interact with both theretinoblastoma tumor suppressor protein and the cytokinesis regulatory protein,citron-K kinase (Atreya et al., 1998, 2004a). There is evidence that a balance be-tween cellular survival signals and normal proliferative signals is needed for op-timal virus growth (Cooray et al., 2005). Although one must note that molecularevidence for the mechanism(s) of teratogenesis are lacking in humans or animalmodels, these works in cell lines suggest that interactions between rubella proteins(e.g. p90) and with cellular proteins which regulate cell growth may be the reasonwhy rubella virus is a potent human teratogen (reviewed by Atreya et al., 2004b).

Rubella virus establishes a persistent infection in a number of cell types(Chantler et al., 2001). Given the fact that long-term infection is an importantfeature of the primary manifestation of rubella (CRS), the molecular mechanism(s)for establishment of persistence deserves more investigation.

Assembly of virions

Alphaviruses such as Sindbis undergo a rather straightforward assembly process inwhich the glycoproteins E1 and E2 are processed through the Golgi while thenucleocapsids are formed in the cytoplasm, and then nucleocapsids and glycopro-teins interact through a budding process at the plasma membrane (Frey andWolinsky, 1999; Schlesinger, 1999). Rubella virus assembly does not seem to haveseparate pathways for glycoproteins and nucleocapsids, possibly because the Cprotein is membrane bound via the E2 signal sequence (Suomalainen et al., 1990).

The three SP of rubella virus are translated from the subgenomic RNA into apolyprotein, which is cleaved into the individual proteins by signalase (Frey andWolinsky, 1999). The E1 and E2 proteins undergo straightforward glycosylationprocesses in which the site of N-glycosylation is the lumen of the endoplasmicreticulum. The site of O-glycosylation of E2 has not been determined, but the E2protein contains sequences for localization in the Golgi; association of E1 with E2is required for E1 to leave the endoplasmic reticulum (Yang et al., 1998). The Cprotein must associate with viral RNA at some point during virion assembly.Assembly (and disassembly) of viral nucleoprotein and thus viral replication isregulated by phosphorylation and dephosphorylation of the C protein; phosphor-ylation inhibits nucleocapsid formation (Law et al., 2003). A 29-nt segment near the50 end of the rubella virus genomic RNA is essential for capsid binding to the RNA(Liu et al., 1996). Some viral assembly occurs at the plasma membrane, but mostassembly occurs at intracellular membranes (Chantler et al., 2001). Assembly ofrubella virus virions occurs in the Golgi and maturation of virions involving con-densation of the nucleocapsids has been observed in the Golgi (Risco et al., 2003).Furthermore, Golgi stacks were also detected near cytopathic vacuoles containingRNA-replication complexes, suggesting an integrated RNA replication andassembly process; as mentioned above, the C protein is known to influence RNAreplication (Law et al., 2003; Chen and Icenogle, 2004). Rubella virus virions are


released from the infected cell in a cell-type-dependent manner; more infectiousparticles are released from Vero cells than from BHK cells (discussed in Risco et al.,2003). Since glycosylation, transport, and assembly of rubella proteins all occur inthe Golgi, it is difficult to sort out factors essential for each process (discussed inYao and Gillam, 2000).

Molecular epidemiology of rubella

Despite the fact that rubella virus has a similar genomic organization to thealphaviruses, the phylogenetic relationship between the Rubivirus genus and theAlphavirus genus are distant and the evolutionary events connecting rubella viruswith alphaviruses are apparently complicated (Gorbalenya et al., 1991; Kooninet al., 1992; Weaver et al., 1993; Frey and Wolinsky, 1999). However, a number ofstudies have described currently (or recently) circulating rubella viruses in the worldand these viruses fall into a relatively small number of phylogenetic groups (Katowet al., 1997; Zheng et al., 2003a). These rubella viruses group into two virus clades(formerly called genotypes), which differ in nt sequence by about 8–10% (Katow etal., 1997; Frey et al., 1998). Each clade consists of a number of genotypes (Frey etal., 1998; Zheng et al., 2003a). A systematic nomenclature proposed at a WHOmeeting in September 2004 consists of two clades (1 and 2) and seven genotypes(designated by clade and a letter designation, e.g. 1C) and three provisionalgenotypes (1a, 1 g, and 2c) (Fig. 4) (WHO, 2005).

Molecular epidemiology has supported control and elimination of vaccine pre-ventable diseases, e.g. polio and measles. For example, the discovery that vaccine-derived poliovirus recombinants can circulate and cause poliomyelitis contributedto understanding a poliomyelitis outbreak in Hispaniola that started in 2000 (Kewet al., 2002). As another example, in 2001, an outbreak of measles in Venezuela andColumbia was caused by an imported measles virus genotype D9 rather than D6,which had caused previous outbreaks in the region (CDC, 2003a,b).

There is sufficient variability in currently circulating rubella viruses to allow themolecular epidemiology of rubella viruses to support rubella control and elimina-tion activities. For example, clade 2 viruses have not been found circulating in theAmericas, and thus clade 2 viruses found in this region are considered importations(WHO, 2000). Some rubella virus genotypes are geographically restricted, allowingsporadic cases caused by these genotypes in other locations to be considered im-portations (CDC, 2005a). A shift in genotype from 1a and closely related viruses toa mixture of other genotypes (1B–E) occurred in the United States around 1981(authors’ unpublished data). The genotypes in the United States after 1981 areconsistent with rubella viruses circulating in neighboring countries between 1981and 2001. This consistency is even clearer for viruses isolated between 1996 and2000. This molecular evidence was used to support the conclusion that indigenousrubella and CRS have been eliminated from the United States (Reef et al., 2002;CDC, 2005b; Icenogle et al., in press).


Although the phylogeny and molecular epidemiology of rubella virus is becom-ing clear, the best operational taxonomic unit(s) for the molecular epidemiology ofrubella viruses remains to be determined for some situations. For example, considerrubella virus molecular epidemiology in populations with low vaccine coverage.Multiple genotypes of rubella virus can co-circulate in such populations (Donadioet al., 2003; Zheng et al., 2003b). Rubella viruses in such populations might be veryheterogeneous, reflecting many epidemics, importations, etc.; different virus lineages,within a genotype, may be useful to track. When tracking viruses within a genotype,extensive sequence information (>3000 nts) might be needed to fully understand therubella epidemiology, as has been the case for poliovirus (Liu et al., 2003).

Fig. 4 Phylogenetic tree comparing rubella viruses. A phylogenetic tree using the nucleotide sequence of

the entire structural protein-coding region, nucleotides 6481–9741, for viruses representing the indicated

genotypes is shown. Rubella viruses are currently classified in 2 clades and 10 genotypes designated 1a, 1B,

1C, 1D, 1E, 1F, 1 g, 2A, 2B, and 2c; genotypes which are still provisional are indicated by lower case

letters. Nucleotide sequences from the entire structural protein region are not yet available for provisional

genotype 2c. The tree shown was created using the Bayesian inference program, MRBAYES. The settings

used for the tree were 100,000 ngens; samplefreq, 100; nchains, 4; burnin, 10. Clade credibility values for

the tree are high (80 or above) for all nodes except the ra27 us v (RA27/3 vaccine virus) node.


General comments about the molecular virology of rubella virus

Rubella virus is the most important human pathogen in the Togavirus family, stillcausing at least 100,000 CRS cases in the world each year (Cutts and Vynnycky,1999; WHO, 2000). It is a very potent infectious teratogenic agent. Rubella virus wasintensely studied during the development of rubella vaccines and rubella control indeveloped countries. Indeed, rubella control and elimination from the United Stateswas a monumental and rapid achievement (CDC, 2005b). Now however, thealphaviruses are more heavily studied largely because these viruses are simply easierto grow in quantity. It is unfortunate, given modern techniques, that more currentattention is not given to the molecular virology of rubella virus, an important humanpathogen, which continues to circulate uncontrolled in much of the world. Thereare some intriguing molecular virology problems yet to be solved that are specific tothe Rubivirus genus (e.g. molecular pathogenesis and assembly/disassembly).

References

Atreya CD, Kuldarni S, Mohan KVK. Rubella virus P90 associates with the cytokinesis

regulatory protein Citron-K kinase and the viral infection and constitutive expression of

P90 protein both induce cell cycle arrest following S phase in cell culture. Arch Virol

2004a; 179: 779.

Atreya CD, Lee NS, Forng R-Y, et al. The rubella virus putative replicase interacts with the

retinoblastoma tumor suppressor protein. Virus Genes 1998; 16: 177.

Atreya CD, Mohan KVK, Kulkarni S. Rubella virus and birth defects: molecular insights

into the viral teratogenesis at the cellular level. Birth Defects Res A Clin Mol Teratol

2004b; 70: 431.

Beatch MD, Everitt JC, Law LJ, Hobman TC. Interactions between rubella virus capsid and

host protein p32 are important for virus replication. J Virol 2005; 79: 10807.

CDC (Centers for Disease Control and Prevention). Public health dispatch: absence of

transmission of the d9 Measles virus—region of the Americas, November 2002–March

2003. Morb Mortal Wkly Rep 2003a; 52: 228.

CDC (Centers for Disease Control and Prevention). Progress toward measles elimination—

region of the Americas, 2002–2003. Morb Mortal Wkly Rep 2003b; 53: 304.

CDC (Centers for Disease Control and Prevention). Global measles and rubella laboratory

network, January 2004–June2005. Morb Mortal Wkly Rep 2005a; 54: 1100.

CDC (Centers for Disease Control and Prevention). Achievements in public health: elim-

ination of rubella and congenital rubella syndrome—United States 1969–2004. Morb

Mortal Wkly Rep 2005b; 54: 279.

Chantler J, Wolinsky JS, Tingle A. Rubella virus. In: Fields Virology (Fields BN, Knipe

DM, Howley PM, editors). Philadelphia: Lippincott: Williams and Wilkins; 2001; p. 963.

Chaye H, Chong P, Tripet B, et al. Localization of the virus neutralizing and hemagglutinin

epitopes of E1 glycoprotein of rubella virus. Virology 1992; 189: 483.

Chen J-P, Strauss JH, Strauss EG, Frey TK. Characterization of the rubella virus non-

structural protease domain and its cleavage site. J Virol 1996; 70: 4707.


Chen M-H, Frey TK. Mutagenic analysis of the 30 cis-acting elements of the rubella virus

genome. J Virol 1999; 73: 3386.

Chen M-H, Frolov I, Icenogle JP, Frey TK. Analysis of the 30 cis-acting elements of rubella

virus by using replicons expressing a puromycin resistance gene. J Virol 2004; 78: 2553.

Chen M-H, Icenogle JP. Rubella virus capsid protein modulates viral genome replication

and virus infectivity. J Virol 2004; 78: 4314.

Cooper LZ. The history and medical consequences of rubella. Rev Infect Dis 1985; 7(Suppl

1): S2.

Cooray S, Jin L, Best JM. The involvement of survival signaling pathways in rubella-virus

induced apoptosis. Virol J 2005; 2: 1.

Cordoba P, Lanoel A, Grutadauria S, Zapata M. Evaluation of antibodies against a rubella

virus neutralizing domain for determination of immune status. Clin Diagn Lab Immunol

2000; 7: 964.

Cutts FT, Vynnycky E. Modelling the incidence of congenital rubella syndrome in devel-

oping countries. Int J Epidemiol 1999; 28: 1176.

Dominguez G, Wang C-Y, Frey TK. Sequence of the genome RNA of rubella virus: ev-

idence for genetic rearrangement during togavirus evolution. Virology 1990; 177: 225.

Donadio FF, Siqueira MM, Vyse A, et al. The genomic analysis of rubella virus detected

from outbreak and sporadic cases in Rio de Janeiro state, Brazil. J Clin Virol 2003; 27: 205.

Forng RY, Atreya CD. Mutations in the retinoblastoma protein-binding LXCXE motif of

rubella virus putative replicase affect virus replication. J Gen Virol 1999; 80: 327.

Forng RY, Frey TK. Identification of the rubella virus nonstructural proteins. Virology

1995; 206: 843.

Frey TK. Molecular biology of rubella virus. Adv Virus Res 1994; 44: 69.

Frey TK, Abernathy ES, Bosma TJ, et al. Molecular analysis of rubella virus epidemiology

across three continents, North America, Europe, and Asia, 1961–1997. J Infect Dis 1998;

178: 642.

Frey TK, Wolinsky JS. Rubella virus (Togaviridae). In: Encyclopedia of Virology (Granoff

A, Webster RG, editors). San Diego: Academic Press; 1999; p. 1592.

Frolov I, Schlesinger S. Translation of Sindbis virus mRNA: analysis of sequences down-

stream of the initiating AUG codon that enhance translation. J Virol 1996; 70: 1182.

Gibbons DL, Vaney M-C, Roussel A, et al. Conformational change and protein–protein

interactions of the fusion protein of Semliki Forest virus. Nature 2004; 427: 320.

Gorbalenya AE, Koonin EV, Lai MM-C. Putative papain-related thiol proteases of positive-

strand RNA viruses. Identification of rubi- and aphthovirus proteases and delineation of a

novel conserved domain associated with proteases of rubi-, alpha- and coronaviruses.

FEBS Lett 1991; 288: 201.

Griffin DE. Aphaviruses. In: Fields Virology (Fields BN, Knipe DM, Howley PM, editors).

Philadelphia: Lippincott: Williams and Wilkins; 2001; p. 917.

Gros C, Wengler G. Identification of an RNA-stimulated NTPase in the predicted helicase

sequence of the Rubella virus nonstructural polyprotein. Virology 1996; 217: 367.

Hemphill ML, Forng RY, Abernathy ES, Frey TK. Time course of virus-specific macro-

molecular synthesis during rubella virus infection in Vero cells. Virology 1988; 162: 65.

Hobman TC, Lemon HF, Jewell K. Characterization of an endoplasmic reticulum retention

signal in the rubella virus E1 glycoprotein. Virology 1997; 71: 7670.

Hobman TC, Lundstrom ML, Mauracher CA, et al. Assembly of rubella virus structural

proteins into virus-like particles in transfected cells. Virology 1994; 202: 574.


Ho-Terry L, Cohen A, Tedder RS. Immunological characterisation of rubella virion po-

lypeptides. J Med Microbiol 1984; 17: 105.

Icenogle JP, Frey TK, Abernathy E, Reef SE, Schnurr D, Stewart JA. Genetic analysis of

rubella viruses found in the United States between 1966 and 2004: evidence that indig-

enous rubella viruses have been eliminated. Clin Infect Dis 2006; 43(suppl 3).

Kamer G, Argos P. Primary structural comparison of RNA-dependent polymerases from

plant, animal and bacterial viruses. Nucleic Acids Res 1984; 12: 7269.

Katow S, Minahara H, Fukushima M, Yamaguchi Y. Molecular epidemiology of rubella by

nucleotide sequences of the rubella virus E1 gene in three East Asian countries. J Infect

Dis 1997; 176: 602.

Kee S-H, Cho E-J, Song J-W, et al. Effects of endocytosis inhibitory drugs on rubella virus

entry into VeroE6 cells. Microbiol Immunol 2004; 48: 823.

Kew O, Morris-Glasgow V, Landaverde M, et al. Outbreak of poliomyelitis in Hispaniola

associated with circulating type 1 vaccine-derived poliovirus. Science 2002; 296: 356.

Koonin EV, Gorbalenya AE, Purdy MA, et al. Computer-assisted assignment of functional

domains in the nonstructural polyprotein of hepatitis E virus: delineation of an additional group

of positive-strand RNA plant and animal viruses. Proc Natl Acad Sci USA 1992; 89: 8259.

Kuhn RJ, Zhang W, Rossmann MG, et al. Structure of dengue virus: implications for

flavivirus organization, maturation, and fusion. Cell 2002; 108: 717.

Law LM, Everitt JC, Beatch MD, et al. Phosphorylation of rubella virus capsid regulates its

RNA binding activity and virus replication. J Virol 2003; 77: 1764.

Lee J-Y, Bowden DS. Rubella virus replication and links to teratogenicity. Clin Microbiol

Rev 2000; 13: 571.

Lee J-Y, Hwang D, Gillam S. Dimerization of rubella virus capsid protein is not required for

virus particle formation. Virology 1996; 216: 223.

Lee J-Y, Marshall JA, Bowden DS. Characterization of rubella virus replication complexes

using antibodies to double-stranded RNA. Virology 1994; 200: 307.

Lee J-Y, Marshall JA, Bowden DS. Localization of rubella virus core particles in Vero cells.

Virology 1999; 265: 110.

Lemm JA, Rumenapf T, Strauss EG, et al. Polypeptide requirements for assembly of func-

tional Sindbis virus replication complexes: a model for the temporal regulation of minus-

and plus-strand RNA synthesis. EMBO J 1994; 13: 2925.

Liang Y, Gillam S. Mutational analysis of the rubella virus nonstructural polyprotein and its

cleavage products in virus replication and RNA synthesis. J Virol 2000; 74: 5133.

Liang Y, Gillam S. Rubella virus RNA replication is cis-preferential and synthesis of neg-

ative- and positive-strand RNAs is regulated by the processing of nonstructural protein.

Virology 2001; 282: 307.

Liang Y, Yao J, Gillam S. Rubella virus nonstructural protein protease domains involved in

trans- and cis-cleavage activities. J Virol 2000; 74: 5412.

Liu H-M, Zheng D-P, Zhang L-B, et al. Serial recombination during circulation of type 1

wild-vaccine recombinant polioviruses in China. J Virol 2003; 77: 10994.

Liu Z, Yang D, Qui Z, et al. Identification of domains in rubella virus genomic RNA and

capsid protein necessary for specific interaction. J Virol 1996; 70: 2184.

Magliano D, Marshall JA, Bowden DS, et al. Rubella virus replication complexes are virus-

modified lysosomes. Virology 1998; 240: 57.

Marr LD, Wang C-Y, Frey TK. Expression of the rubella virus nonstructural protein ORF

and demonstration of proteolytic processing. Virology 1994; 198: 586.


Modis Y, Ogata S, Clements D, Harrison SC. Structure of the dengue virus envelope protein

after membrane fusion. Nature 2004; 427: 313.

Myles KM, Pierro DJ, Olson KE. Deletions in the putative cell receptor-binding domain of

Sindbis virus strain MRE16 E2 glycoprotein reduce midgut infectivity in Aedes aegypti. J

Virol 2003; 77: 8872.

Nakhasi HL, Ramanujam M, Atreya CD, et al. Rubella virus glycoprotein interaction with

the endoplasmic reticulum calreticulin and calnexin. Arch Virol 2001; 146: 1.

Pappas CL, Tzeng WP, Frey TK. Evaluation of cis-acting elements in the rubella virus

subgenomic RNA that play a role in its translation. Arch Virol 2006; 151: 327.

Paredes AM, Ferreira D, Horton M, et al. Conformational changes in Sindbis virions re-

sulting from exposure to low pH and interactions with cells suggest that cell penetration

may occur at the cell surface in the absence of membrane fusion. Virology 2004; 324: 373.


Proc Soc Exp Biol Med 1962; 111: 225.

Perrenoud G, Messerli F, Thierry A-C, et al. A recombinant rubella virus E1 glycoprotein as

a rubella vaccine candidate. Vaccine 2004; 23: 480.

Pogue GP, Cao XQ, Singh NK, Nakhasi HL. 50 sequences of rubella virus RNA stimulate

translation of chimeric RNAs and specifically interact with two host-encoded proteins. J

Virol 1993; 67: 7106.

Pugachev KV, Abernathy ES, Frey TK. Improvement of the specific infectivity of the rubella

virus (RUB) infectious clone: determinants of cytopathogenicity induced by RUB map to

the nonstructural proteins. J Virol 1997; 71: 562.

Pugachev KV, Frey TK. Effects of defined mutations in the 50 nontranslated region of

rubella virus genomic RNA on virus viability and macromolecule synthesis. J Virol 1998a;

72: 641.

Pugachev KV, Frey TK. Rubella virus induces apoptosis in culture cells. Virology 1998b;

250: 359.

Pugachev KV, Tzeng W-P, Frey TK. Development of a rubella virus vaccine expression

vector: use of a picornavirus internal ribosome entry site increases stability of expression. J

Virol 2000; 74: 10811.

Qiu Z, Tufaro F, Gillam S. The influence of N-linked glycosylation on the antigenicity and

immunogenicity of rubella virus E1 glycoprotein. Virology 1992; 190: 876.

Qui Z, Yao J, Cao H, Gillam S. Mutations in the E1 hydrophobic domain of rubella virus

impair virus infectivity but not virus assembly. J Virol 2000; 74: 6637.

Ramanujam M, Hofmann J, Nakhasi HL, Atreya CD. Effect of site-directed asparagine to

isoleucine substitutions at the N-linked E1 glycosylation sites on rubella virus viability.

Virus Res 2001; 81: 151.

Reef SE, Frey TK, Theall K, et al. The changing epidemiology of rubella in the 1990s: on the

verge of elimination and new challenges for control and prevention. JAMA 2002; 287:

466.

Risco C, Carrascosa JL, Frey TK. Structural maturation of rubella virus in the Golgi

complex. Virology 2003; 312: 261.

Schlesinger MJ. Sindbis and Semliki Forest viruses (Togaviridae). In: Encyclopedia of Vi-

rology (Granoff A, Webster RG, editors). San Diego: Academic Press; 1999; p. 1656.

Strauss JH, Strauss EG. Structure and replication of the alphavirus genome. In: The To-

gaviridae and Flaviviridae (Schlesinger S, Schlesinger MG, editors). New York: Plenum

Press; 1986; p. 35.


Suomalainen M, Garoff H, Baron MD. The E2 signal sequence of rubella virus remains part

of the capsid protein and confers membrane association in vitro. J Virol 1990; 64: 5500.

Tzeng W-P, Chen M-H, Derdeyn CA, Frey TK. Rubella virus DI RNAs and replicons:

requirement for nonstructural proteins acting in cis for amplification by helper virus.

Virology 2001; 289: 63.

Tzeng W-P, Frey TK. Mapping the rubella virus subgenomic promoter. J Virol 2002; 76:

3189.

Tzeng W-P, Frey TK. Complementation of a deletion in the rubella virus p150 nonstructural

protein by the viral capsid protein. J Virol 2003; 77: 9502.

Wang C-Y, Dominguez G, Frey TK. Construction of rubella virus genome-length cDNA

clones and synthesis of infectious RNA transcripts. J Virol 1994; 68: 3550.

Wang X, Gillam S. Mutations in the GDD motif of rubella virus putative RNA-dependent

RNA polymerase affect virus replication. Virology 2001; 285: 322.

Wang X, Liang Y, Gillam S. Rescue of rubella virus replication-defective mutants using

vaccinia virus recombinant expressing rubella virus nonstructural proteins. Virus Res

2002; 86: 111.

Weaver SC, Hagenbaugh A, Bellew LA, et al. A comparison of the nucleotide sequences of

eastern and western equine encephalomyelitis viruses with those of other alphaviruses and

related RNA viruses. Virology 1993; 197: 375.

Weller TH, Neva FA. Propagation in tissue culture of cytopathic agents from patients with

rubella-like illness. Proc Soc Exp Biol Med 1962; 111: 215.

WHO (World Health Organization). Preventing congenital rubella syndrome. Wkly

Epidemiol Rec 2000; 75: 290.

WHO (World Health Organization). Standardization of the nomenclature for genetic char-

acteristics of wild-type rubella viruses. Wkly Epidemiol Rec 2005; 80: 126.

Wolinsky JS, Sukholutsky E, Moore WT, et al. An antibody- and synthetic peptide-defined

rubella virus E1 glycoprotein neutralization domain. J Virol 1993; 67: 961.

Yang D, Hwang D, Qui Z, Gilliam S. Effects of mutations in the rubella virus E1 glyco-

protein on E1–E2 interaction and membrane fusion activity. J Virol 1998; 72: 8747.

Yao J, Gillam SA. single-amino-acid substitution of a tyrosine residue in the rubella virus E1

cytoplasmic domain blocks virus release. J Virol 2000; 74: 3029.

Zhang W, Mukhopadhyay S, Pletnev SV, et al. Placement of the structural proteins in

Sindbis virus. J Virol 2002; 76: 11645.

Zheng D-P, Frey TK, Icenogle J, et al. Global distribution of rubella virus genotypes. Emerg

Infect Dis 2003a; 9: 1523.

Zheng D-P, Zhou YM, Zhao K, et al. Characterization of genotype II Rubella virus strains.

Arch Virol 2003c; 148: 1835.

Zheng D-P, Zhu H, Revello MG, et al. Phylogenetic analysis of rubella virus isolated during

a period of epidemic transmission in Italy, 1991–1997. J Infect Dis 2003b; 187: 1587–1597.


Rubella Viruses



DOI 10.1016/S0168-7069(06)15002-8

19

Chapter 2

Clinical Features: Post-Natally AcquiredRubella

Jangu E. BanatvalaKing’s College London Medical School at Guy’s, King’s College and St Thomas’Hospitals, London, SEI 7EH, UK

Introduction

Infection is transmitted via the aerosol route. Infected persons may excrete highconcentrations of virus (e.g. Z106 TCID 50/0.1ml) although studies on volunteersexcreting a rubella vaccine strain (RA27/3) showed that variations in titre as largeas a 1000-fold may occur even over a 4–6-h period (Harcourt et al., 1979). Nev-ertheless, in contrast with infections such as varicella and measles, close and pro-longed contact is usually necessary for rubella to be transmitted to susceptiblecontacts.

Following attachment to receptors likely to be localised in the buccal mucosaand nasopharynx, virus spreads to and replicates in the lymphoid tissue in thenasopharynx and upper respiratory tract. Systemic infection then follows, involv-ing many organs, including the placenta.

Clinical features

Fig. 1 illustrates the relationship between the clinical and virological features ofpost-natally acquired rubella. Although the incubation period is usually around13–20 days, this relates to the interval between exposure and onset of rash.

However, lymphadenopathy, which is usually the first clinical manifestation ofinfection, may precede the onset of rash for up to a week, and persist for up to10–14 days after the rash has disappeared. The post-auricular and cervical lymph

dx.doi.org/10.1016/S0168-7069(06)15002-8.3d

nodes are usually involved; in adults the lymph nodes are often tender, but less soamong children.

In childhood constitutional symptoms are usually mild or absent, and the onsetis usually abrupt with the appearance of rash. Adults may develop such prodromalfeatures as fever and malaise, this being associated with viraemia. This is usually of4–5 days duration and terminates with the development of neutralising antibodies,which develop at about the same time as the rash appears. Viraemia is bothlymphocytic and cell free.

The rash, which is at first discreet and presents as pinpoint macular lesions,appears initially on the face and then spreads rapidly to the trunk and limbs.Lesions may coalesce but the rash seldom persists for longer than 3–4 days; inmany cases the rash is fleeting. However, rubella may present atypically with anevanescent rash and, in up to 25% of cases, infection may be subclinical.

Since the appearance of rash coincides with the development of humoral im-mune responses, it has been suggested that this is mediated by antigen–antibodycomplexes. Rubella virus has been isolated from areas of the skin with rash as wellas rash-free areas (Heggie, 1978).

Neut . & HAI antibody

SRH antibody

Specific IgG (EIA)

Specific IgM (EIA)

Pharynx

Blood

Stool

Urine

Rash

Fever

Arthralgia

Lymphadenopathy

Sero

logy

Vir

us I

sola

tion

Clin

ical

Fea

ture

s

Days after exposure

393837

30282624222018161412108642

30282624222018161412108642

Fig. 1 Relation between clinical and virological features of post-natally acquired rubella.

J.E. Banatvala20

Virus may be recovered from nasopharyngeal secretions for about a week be-fore the onset of rash and for a similar period, but sometimes longer, after the rashhas disappeared. Virus may also be detected in the blood, stool and urine, but theseare not favoured sites for recovering virus for diagnostic purposes (see Chapter 3).

Differential diagnosis

Since rubella may present atypically or with non-specific symptoms and signs thatmay be caused by other viruses which do not have a teratogenic potential, it isimportant that the clinical diagnosis be confirmed by laboratory investigations,particularly during pregnancy. Rubella-like rashes may be induced by such virusesas parvovirus B19, enteroviruses, adenoviruses and such arboviruses as Ross Riverand Chikungunya. Arthralgia may be a feature of some of these infections, e.g.parvovirus B19, Ross River and Chikungunya. In developing countries, in additionto the above infections, rubella may also be readily confused clinically with dengueand measles, and differentiating these infections is of importance during surveil-lance programmes involved in rubella elimination (see Chapter 4). Table 1 showsthe geographical distribution of some virus infections which may be difficult orimpossible to distinguish from rubella.

Complications

Joint involvement is the commonest complication of naturally acquired rubella aswell as following rubella vaccination. This usually develops as rash subsides and,although uncommon among males and pre-pubertal females, may occur in up to50% of post-pubertal females. Symptoms vary in severity, ranging from a transientarthralgia with joint stiffness, to a frank arthritis with pain, limitation of movementand swelling. The finger joints, wrists, knees and ankles are most frequently affected.Generally, symptoms persist for 3–4 days, but may last for up to a month and evenoccasionally longer, exhibiting a fluctuating course. The pathogenesis is the subjectof some debate. Hormonal factors are likely to be involved, since in addition tooccurring commonly in post-pubertal females, studies among vaccinees have shownthat joint symptoms are likely to develop within 7 days of the onset of the menstrualcycle (Harcourt et al., 1979). Virus has been isolated from the synovial fluid ofpatients with naturally acquired and vaccine-induced arthritis (Fraser et al., 1983;Best and Banatvala, personal communication) and since symptoms develop whenrash subsides and humoral antibodies appear, immune complexes may play a part inpathogenesis. Some investigators have suggested that persistent rubella infectionmay be associated with chronic arthritis (Chantler et al., 1985; Smith et al., 1987)but a study in which synovial fluid and biopsies of synovial membrane from adultswere examined by RT-PCR as well as by virus recovery in cell culture, failed toshow an association with chronic inflammatory joint disease (Bosma et al., 1998).

Clinical Features: Post-Natally Acquired Rubella 21

Table 1

Differential diagnosis of postnatal rubella in different geographical regions

Geographical distribution Key features

Africa Asia Australia Europe North

America

Central

America

South

America

Pacific

Virus infection

Rubella + + + + + + + +

Parvovirus B19 + + + + + + + + Erythema infectiosum

Human herpes viruses 6 and 7 + + + + + + + + Exanthem subitum,

mainlyo2 years

Measles + + + + + + + + Prodrome with cough,

conjunctivitis, coryza

Enteroviruses + + + + + + + + Echovirus 9,

Coxsackie A9 most

frequent

Dengue + + + � � + + + Joint and back pain,

haemorrhagic

complications in

children

West Nile fever + + � + + � � � Joint pain

Chickungunya + + � � � � � � Joint pain

Ross River � � + � � � � + Joint pain

Sindbis + + + + � � � � Joint pain

Source: Reproduced from Banatvala and Brown (2004).

J.E

.B

an

atva

la22

In cases in which joint symptoms persist or relapse occurs, parasthesiae, some-times associated with carpal tunnel syndrome, may develop.

CNS complications are rare. Post-infectious encephalopathy or encephalomy-elitis occur in 1 : 5000 to 1 : 10,000 cases, symptoms usually developing about aweek after the onset of rash (reviewed by Frey, 1997). In general, recovery occurswithin 7–30 days, although the course may be prolonged death is unusual. Symp-toms are similar to those among patients with other viral encephalitidies, althoughpatients may not be febrile. There is a lymphocytic pleocytosis but intrathecalrubella virus antibody synthesis has not been reported, although rubella virus RNAhas been detected in the CSF by reversed transcription PCR (Date et al., 1995).Although death is very rare, post-mortem studies show that, in contrast with otherpost-infectious encephalitidies, demylination is not present, which suggests thatimmune mechanisms are not involved in its pathogenesis.

Gullain–Barre syndrome, although an even rarer complication of rubella, hasbeen reported (Atkins and Esmonde, 1991) and consequently it is worth investi-gating patients lest they have experienced recent infection by rubella, bearing inmind that many adults experience subclinical infections.

Major haematological abnormalities are unusual. Haemolytic anaemia hasbeen reported during outbreaks in Japan (Ueda et al., 1985). As with other ex-anthemata, a transient depression of thrombocytes may occur, but a purpuric rashis uncommon, occurring in about 1 : 1500 cases (Steen and Torp, 1956; Ueda et al.,1985). It is possible that thrombocytopenia may result from enhanced clearance ofplatelets by the reticuloendothelial system, which may be the result of an auto-immune phenomenon (Rand and Wright, 1998).

Unlike measles and varicella, those with impaired cell-mediated immunity, havenot been reported to suffer unduly severely from rubella. Thus although theAmerican Academy of Pediatrics, while not recommending the use of live-atten-uating vaccines to severely immunocompromised HIV-infected children, expressthe view that MMR vaccine should not be withheld from those with no evidence ofimmunosuppression or those who have only a moderate amount of immune sup-pression (American Academy of Pediatrics, 1999).

There has, however, been a report of a child being inadvertently vaccinated inremission from acute lymphoblastic leukaemia who developed chronic arthritis,and persistent viraemia, despite developing high antibody titres (Geiger et al.,1995).

Congenitally acquired infection

Pathogenesis of congenital rubella

If maternal rubella is acquired during the first trimester of pregnancy, particularlybefore 8 weeks, this is likely to result in a generalised and persistent infection withmulti-system disease.


The severity of infection following maternal rubella in the first trimester, reflectsthe inefficiency of the placental barrier as well as the inability of the fetus to mountan effective immune response to eliminate rubella virus.

The mechanism by which fetal damage occurs is probably multi-factorial, beingthe result of rubella-induced cellular damage and the effect of the virus on dividingcells. The fetus is at risk during the period of maternal viraemia at which timeplacental infection is likely to occur. Tondury and Smith (1966) in their classicalhistopathological study showed that following maternal rubella in early pregnancy,the presence of focally distributed areas of necrosis in the epithelium of the chor-ionic villae and in the endothelial cells of its capillaries. These endothelial cellsappeared to be desquamated into the lumen of the vessels, suggesting that virus istransported into the fetal circulation in the form of rubella-infected endothelial cell‘‘emboli’’ which may result in the blockage of small blood vessels and, if trans-ported into the fetal circulation, result in infection and damage occurring in variousfetal organs. The most extensive histological changes occur in early pregnancy, i.e.before fetal defence mechanisms are sufficiently mature to be activated. Thus, amajor characteristic of rubella embryopathy in the early weeks of gestation is thepresence of cell necrosis in the absence of an inflammatory response.

In addition to cellular damage, there is evidence of retardation in cellular rep-lication, thus infected cells have a reduced lifespan (Rawls and Melnick, 1966),indeed it has been shown that the organs of rubella-infected infants are smaller andcontain a subnormal number of cells when compared with those of healthy infants(Naeye and Blanc, 1965). Although virus may spread from infected cells to progenycells during division, resulting in infected clones of cells (Woods et al., 1966; Rawlset al., 1968), only discreet foci can be detected in a rubella-infected conceptus. Thenumber of infected cells is small (about �0.1%) and such infected foci are protectedfrom maternal and fetal immune responses.

During the second trimester, fetal immune responses gradually mature andsome transfer of maternal rubella-specific IgG occurs, whereas prior to this, pla-cental transfer is inefficient and the fetus is consequently defenceless against in-fection. Fetal immunoglobulin secreting plasma cells increase progressively duringthe second trimester and may be detected after about 15 weeks of gestation (re-viewed by Webster, 1998).

Cell-mediated immune responses are involved in the elimination of cells in-fected by such enveloped viruses as rubella. Cytotoxic T cells, natural killer cellsand monocytes, together with the secretion of lymphokines, are involved, but suchresponses like humeral responses do not mature until the second trimester. Al-though T cells may be detected as early as the 12th week of gestation, adult ex-pression is delayed for another 5–6 weeks (Lobach et al., 1985), and lymphokineproduction is immature. Thus, during the second trimester and subsequently, thefetus is afforded protection by the progressive development of immune responses,although in vitro studies indicate that fetal organs are still susceptible to rubella(Best et al., 1968). However, rubella virus itself can have an adverse effect onimmune responses, since a feature of the congenital rubella syndrome (CRS)

J.E. Banatvala24

neonatally, is the presence of a dysgammaglobulinaemia, with either an increasedor a decreased level of immunoglobulins.

Infants with congenitally acquired infection may develop manifestations of late-onset disease in later childhood or as adults. The endocrine organs, the eyes, hear-ing and the CNS may be affected and behavioural problems, including autism, mayoccur. Autoimmune phenomena may be responsible for endocrine and CNS dis-turbances. Rubella virus has been recovered from a cataract removed from a 3-year-old child (Menser et al., 1967a), rubella antigen from the thyroid of a 5-year-old child with Hashimoto’s disease (Ziring et al., 1977) and virus has been recov-ered from the brain of a 12-year-old child developing rubella panencephalitis(Cremer et al., 1975; Weil et al., 1975).

Risks of congenital rubella

Following maternal rubella in the first trimester

The outcome of maternal rubella infection in the first trimester is very varied, asFig. 2 below illustrates. However, following virologically confirmed rubella in thefirst trimester, the fetus is almost invariably infected, and in about 80–85% of casesaffected manifesting a wide spectrum of anomalies (Miller et al., 1982; Wolinsky,1996).

Fig. 3 correlates the range of congenital anomalies with gestational age, among376 infants with congenitally acquired rubella following the extensive 1964 rubellaepidemic in the USA. Cardiac and eye defects occurred during the critical phase oforganogenesis in the first 8 weeks of pregnancy, whereas retinopathy and deafnesswere more evenly distributed through the first 16 weeks of gestation.

Studies on autopsy material obtained from infants dying in early infancy fromsevere and generalised infections, show that virus may be recovered from virtuallyevery organ of the body. Neonatally, virus may be recovered from nasopharyngeal

Fetalinfection

Abortion/still birth

Congenitalmalformation

Congenitaldisease

Disease ininfancy

Normalinfant

Normalchild

Lateonset disease

Fig. 2 Possible outcomes of congenital rubella.


secretions, urine, stools and tears. Among infants with congenitally acquired ru-bella, most are excreting virus at birth but by 3 months, and 9–12 months, theproportion has declined to 50–60% and 10%, respectively. High titres of virus areexcreted at birth and such infants may readily transmit infection to susceptiblecontacts. The mechanism by which rubella persists throughout gestation and for alimited period during the first year of life, remains to be clearly established. Possiblemechanisms may involve defective cell-mediated immunity. Thus, Buimovici-Kleinet al. (1979) demonstrated diminished lymphoproliferative responses tophytohaemagglutin and rubella antigen and diminished interferon synthesis inchildren aged 1–12, this being related to the gestational age at which maternalinfection occurred. It was greatest during the first 8 weeks. Failure to mount alymphoproliferative response may be useful diagnostically in determining whetherchildren were infected in utero if being investigated after the first year of life, atwhich time virus excretion and virus-specific IgM responses would no longer bedetectable (O’Shea et al., 1992). Alternatively, as stated above, the possibility that anumber of virus-infected cells result in infected clones which have a limited life spanand when these cells no longer survive, virus excretion terminates (reviewed byBanatvala, 1977; Banatvala and Best, 2000). Selective immune tolerance to the viralE1 protein has also been suggested (Mauracher et al., 1993).

Post-first trimester

In contrast with infants whose mothers acquire rubella in the first trimester, rubellais seldom isolated from infants whose infection is acquired at a later gestationalage. Nevertheless, serological studies confirm that about one-third of infants whose

0

10

20

30

40

50

60

70

80

1 2 3 4 >4month of pregnancy

% o

f in

fan

ts w

ith

sp

ecif

ic d

efec

tsNeonatal purpura

Heart disease

Cataract or glaucoma

Deafness

Neurologic deficit

Fig. 3 Relationship between the gestational age and maternal infection and clinical manifestations of

clinical rubella. With permission from John Wiley & Son.

J.E. Banatvala26

mothers were infected between the 16th and 20th week of pregnancy, were infectedsince rubella-specific IgM is present neonatally (Craddock-Watson et al., 1980).

However, as organogenesis is complete by the end of the first trimester, and fetalimmune responses, which may limit or terminate infection, mature during the sec-ond trimester, infected infants rarely have severe or multiple anomalies.

An analysis of four studies conducted in different countries showed that be-tween 13 and 16, 17 and 20 and after 20 weeks, the proportion with rubella-induceddefects were 16.7%, 15.9% and 1.7%, respectively (Banatvala and Best, 2000). Ingeneral, deafness, which was usually the sole manifestation of infection, and re-tinopathy, which per se does not usually affect vision, were the only anomaliesgenerally associated with maternal rubella after the first trimester. This stresses theimportance of careful assessment of infants with post-first trimester serologicalevidence of intrauterine infection who appear to be unaffected in order to detecthearing defects as early as possible.

Pre-conception

Although there have been occasional reports on maternal infection occurring as longas 21 days prior to conception published prior to 1984 (reviewed by Best and Ban-atvala, 2000), a prospective study conducted in Britain and Germany showed thatpre-conceptional rubella did not result in transmission of rubella to the fetus. Thus,there was no serological or clinical evidence of intrauterine infection in 38 infantswhose mother’s rash appeared before or within 11 days after their last menstrualperiod (LMP), although after an interval of 12 days had elapsed between the LMPand the rash, fetal infection occurred. All of the 10 mothers developing rash 3–6weeks after their LMP resulted in viral transmission to the fetus (Enders et al., 1988).

Clinical features of congenital rubella

In addition to the spectrum of anomalies, which may be apparent at birth or duringthe early months of life, congenitally acquired infection may result in a consid-erably delayed onset of disease manifestations which may not become apparentuntil adolescence or even later. Thus, congenitally acquired rubella should be re-garded as a chronic disease, although it may not always be possible to detect virusin late-onset disease.

The spectrum of manifestations of congenitally acquired rubella is variableaccording to the gestational period during which maternal rubella is acquired.Congenital disease may only affect, for example, hearing, particularly if acquiredpost first trimester. If acquired at this stage or even later during gestation, theremay be serological evidence of infection but without evidence of disease, althoughlong term follow up is advisable lest late onset manifestations occur.

A wide spectrum of congenital anomalies occurs when maternal rubella is ac-quired during early gestation, particularly if acquired during the critical phase for


organogenesis. Hearing, eye and cardiac defects predominate but other anomaliesmay also occur. The term congenital rubella syndrome (CRS) is used for infantswith such anomalies. Some of these are listed in Table 2 together with the time atwhich they are commonly recognised. Early and transient features are self-limiting,but permanent and developmental defects which, although taking months or evenyears before becoming apparent, persist indefinitely. Virtually every organ may be

Table 2

Clinical features of congenital rubella syndrome

Time when signs

commonly

recogniseda

Early transient

features

Permanent

featuresb

Ocular defects

Cataracts (unilateral or bilateral) Early infancy � +

Glaucoma Early infancy � +

Pigmentary retinopathy Early infancy � +

Microphthalmia � +

Iris hypoplasia � +

Cloudy cornea + �

Auditory defects

Sensorineural deafness (unilateral

or bilateral)

Early infancy � +

Cardiovascular defects

Persistent ductus arteriosus Early infancy � +

Pulmonary artery stenosis Early infancy � +

Ventricular septal defect Early infancy � +

Myocarditis + �

CNS defects

Microcephaly Neonatal � +

Psychomotor retardation � +

Meningoencephalitis Neonatal + �

Behavioural disorders � �

Speech disorders

Intrauterine growth retardation + �

Thrombocytopenia, with purpura Neonatal + �

Hepatitis/hepatosplenomegaly Neonatal + �

Bone lesions Neonatal + �

Pneumonitis + �

Lymphadenopathy + �

Diabetes mellitus � +

Thyroid disorders � +

Progressive rubella panencephalitis � +

Source: Reproduced from Banatvala and Brown (2004).aTime when signs seen commonly used to clinically confirm CRS cases.bSome recognised as late as during adolescence or as adults.

J.E. Banatvala28

affected. Fig. 4 shows the characteristic purpuric rash (‘‘blueberrymuffin’’ appear-ance) often apparent at birth in infants whose mothers acquired infection in earlypregnancy and who usually have or develop other clinical features such as car-diovascular, eye or hearing anomalies. This rash is associated with a low plateletcount and usually spontaneously resolves within a few weeks.

Infants aged between 3 and 12 months may develop a chronic rubella-like rash,pneumonitis and diarrhoea, which are sometimes referred to as ‘‘late-onset dis-ease’’. Although there is a high mortality rate, infants often improve with treatmentwith corticosteroids and it has been suggested that immune complexes may beresponsible for this syndrome.

Cardiovascular anomalies

A patent ductus arteriosus (PDA) and septal defects in association with pulmonaryartery and pulmonary valvular stenosis are common and responsible for much of

Fig. 4 Purpuric rash in infant with CRS.


the high perinatal mortality associated with congenital rubella. Fallot’s tetralogymay also occur, but more rarely. Some infants also have evidence of neonatalmyocarditis (Hastreiter et al., 1967). Rubella may also result in damage and pro-liferation of the intimal lining of the arteries, causing obstructive lesions of thelarger- and medium-sized arteries in the systemic and pulmonary circulation. Thismay result in stenosis of coronary, cerebral and renal arteries (Fortuin et al., 1971;Sever et al., 1985); renal obstruction may result in hypertension (Menser et al.,1967b; Menser and Reye, 1974).

Sensorineural hearing loss

Hearing defects of varying degrees of severity (unilateral and bilateral) are the mostcommon manifestations of congenital rubella, which may be present in 70–90% ofcases of children with rubella-associated defects. The burden of deafness due tocongenital rubella has almost certainly been underestimated. Congenital rubella isconsidered to be the most important cause of non-genetic congenitally acquiredhearing loss in countries, which are not implementing rubella vaccination pro-grammes (Smith, 1999). Rubella may at first infect the labyrinth, although clinicalevidence of vestibular malfunction is uncommon. Rubella then passes through thestria vascularis to infect the sensorial cells of the organ of corti, which results inprogressive sensorial hearing loss (Scasso et al., 1998).

Some infants, although having apparently normal hearing in infancy, maysubsequently develop sensoneural deafness, which may increase over time. Theyshould therefore be followed up regularly. There has been a report of sudden onsetdeafness in a child of aged 10 who had apparently normal hearing until then (Severet al., 1985).

Newer methods are now used to assess hearing in infancy, which includesotoacoustic emissions and auditory brainstem responses (Mehl and Thomson,1998). Infants delivered of mothers with a history of maternal rubella shouldtherefore be screened by such techniques in early infancy since hearing defects mayoften be detected much earlier than hitherto in the neonatal period (van Straaten etal., 1996). They will not be detected if it is progressive so caution is needed. Promptrecognition and management is of importance in order to reduce delays in languagedevelopment and social adaptation. Unfortunately the equipment is costly and itsreliability in field settings has yet to be determined. This currently limits its ap-plication in developing countries where congenital rubella defects continue to be amajor problem.

Eye defects

The major defects first described in Gregg’s epoch-making paper in 1941 (Gregg,1941), included cataract, which was usually bilateral and filled the pupillary area,and microphthalmia (Fig. 5). Rubella virus may affect not only the lens, but alsothe retina and ciliary body. Apart from retinopathy, cataracts, which can be

J.E. Banatvala30

unilateral, have been shown to be the most common manifestation of eye defectscaused by congenital rubella. However, they are a relatively rare manifestation ofcongenital infection because the critical time of exposure is small. Retinopathydescribed as ‘‘a salt and pepper appearance in the retina’’, occurs in the majority ofpatients but may be obscured by the presence of cataracts; eye defects may beassociated with strabissmus and nystagmus. There have been reports of late-onsetretinal neovascularisation (Deutman and Grizzard, 1978). Glaucoma, althoughrare in early infancy, may develop in adolescence (Boger, 1980), and may be as-sociated with cataracts and microphthalmia (Givens et al., 1993). Aphakic glau-coma may also occur following cataract aspiration. Corneal abnormalities,including neonatal transient corneal opacities, corneal hydrops and keratoconushave been described (Boger et al., 1981).

Rubella has been reported to be an important cause of visual defects, includingblindness in developing countries, including India (Eckstein et al., 1996), Jamaica(Moriarty, 1988) and Ghana (Lawn et al., 2000).

Central nervous system

Involvement of the brain may result in vascular lesions caused by endothelialdamage as described above as well as direct viral involvement of brain cells(Chantler et al., 1995). MRI scanning shows reduced grey matter and enlargementof the lateral ventricles (Lim et al., 1995). Further studies employing investigationsby ultrasonography and computerised tomographic examination on a smallnumber of patients showed intracranial calcification and linear hyperechogenicityover basal ganglia. Such abnormalities appeared to be predictive of the develop-ment of microcephaly (Chang et al., 1996).

Neonatal manifestations include an encephalitis, some infants exhibiting leth-argy or irritability and seizures. Microcephaly may be present at birth or develop

Fig. 5 Congenital rubella cataract in an infant aged 9 months.


subsequently. On follow up, infants affected by congenital rubella infection maydevelop evidence of varying degrees of psychomotor retardation. Behavioural dis-orders, including autism and a schizophrenia-like disease may develop subse-quently (Cooper, 1985). A higher risk of non-affective psychosis among thoseexposed to rubella in utero during the first trimester, but having normal intelligence,has been described (Brown et al., 2000). This report indicated that these disorderswere unrelated to such factors as deafness, ethnicity or selection bias.

Progressive rubella panencephalitis (PRP) is a rare manifestation of late-onsetdisease and about 50 cases have been described. The outcome is invariably fatal(Frey, 1997). Very rarely, PRP may be a complication of post-natally acquiredrubella. Clinically, the presentation and disease progression is similar to measles-induced subacute sclerosing panencephalopathy. Histopathological studies haveshown enlargement of the lateral ventricles, loss of glial cells and demyelination.The disease probably has an autoimmune aetiology, perhaps as a result of mo-lecular mimicry between viral and host epitopes (Frey, 1997). Serological studiesshown raised antibody titre to rubella in the blood and CSF, with oligoclonal bandsof rubella-specific IgG in the CFS. Rubella virus itself is somewhat illusive and maybe present in a defective form. However, the virus has been isolated following co-cultivation of patients’ brain cells.

Endocrine disorders

Type 1 (insulin-dependent) diabetes mellitus (IDDM) is the most frequent late-onsetmanifestation of congenital rubella after sensorineural hearing loss, first highlightedduring follow-up studies of those affected in the 1939–1941 Australian extensiverubella epidemic. In all, 20% had developed IDDM by the time they were in their30 s (Menser et al., 1978). Follow-up studies on patients with congenital rubellainfection acquired during the 1963–1964 epidemic in the USA reported similarfindings, showing that 12.4% had developed IDDM but also showed that 20% hadpancreatic islet cell cytotoxic surface antibodies (Ginsberg-Fellner et al., 1984). Thepresence of these antibodies, together with a significantly increased HLA-DR3 anddecreased HLA-DR2 haplotype which are associated with autoimmune disorders,are indicative that such mechanisms are involved in the pathogenesis of associatedIDDM with congenital rubella infection (Ginsberg-Fellner et al., 1984).

That autoimmune mechanisms are involved in the pathogenesis of IDDM, issupported by the findings of experimental studies which showed that monoclonalantibodies for rubella virus capsid protein recognise B-cell epitopes on human andrat islet cells (Karounos et al., 1993), and that T-cell clones from patients withcongenital rubella elicit cytotoxic responses to GAD65, a B-cell autoantigen (Ou etal., 1999).

There have been no reports of rubella virus having been detected in pancreatictissue from patients with congenital rubella and this may reflect the shortage ofsuitable autopsy material for investigations. In any case, the precipitating infectioninducing autoimmune phenomena may no longer be present. However, it is

J.E. Banatvala32

possible that viral RNA may persist in greatly reduced genome copies but retainingsome biological activity following clearance of viral infectivity. Although it hasbeen postulated that such virus infections as enteroviruses may be involved in thepathogenesis of IDDM (reviewed by Minor and Muir, 2004) evidence is as yetinconclusive and only rubella virus infections, if congenitally acquired, has beenshown to be involved in the pathogenesis of IDDM among humans. Consequently,further studies using this model may be of importance in elucidating the mech-anisms involved in the pathogenesis of IDDM.

Thyroid disorders, including hypothyroidism, hyperthyroidism and thyroiditis,occur in about 5% of patients with congenital rubella defects. As with IDDM,autoimmune phenomena are likely to be involved, since antibodies to the globulinand microsomal fractures of the thyroid were shown to be present in 23% of the201 deaf patients with congenital rubella (Clarke et al., 1984). Some patients mayhave evidence of both IDDM and thyroid disorders (Floret et al., 1980).

Although growth hormone deficiency may be responsible for poor post-natalgrowth, there have been conflicting reports about the integrity of the hypothalmic/pituitary axis (Preece et al., 1977; Oberfield et al., 1988). Japanese studies showedthat growth retardation correlated with the severity of clinical manifestations atbirth being more severe among those exposed to maternal rubella in the first weekof gestation, particularly among patients with rubella-associated defects who hadcataracts. In contrast, those infected at a later gestational age, although havingsome evidence of growth retardation in infancy and early childhood, had a normalgrowth pattern after starting school (Tokugawa et al., 1986).

Long-term prognosis. The most extensive series of patients studied are those withcongenitally acquired infection occurring during the Australian epidemic in the1940s. Most were of normal stature, only 15% being less than the third percentilefor height (McIntosh and Menser, 1992). Despite most having multiple defects, allof the 40 examined were deaf; they had adjusted socially extremely well. Twentynine (58%) were married, and between them had 51 children. Only one child wasdeaf, but this may well have been hereditary, since his father, who was not affectedby congenital rubella, was also deaf. However, by 60 years of age, this cohort hadan increased prevalence of type 2 diabetes compared with the general population.Follow-up studies had recorded such other disorders as an early menopause (73%)and osteoporosis (13%), all being significantly higher than found in the generalAustralian population (Forrest et al., 2002).These results contrast with follow-up studies on 300 congenital rubella survivors

of the extensive 1963–1964 epidemic in the USA. Only one-third were leading anormal life and the remaining needed care, varying from having to live with rel-atives or requiring 24-h care. Differences between the two studies may, in part, bethe result of advances in clinical management following the 1963–1964 outbreak,reducing mortality, particularly from congenital heart disease, which, during the1940s, was not possible. However, such apparent improvement in prognosis ininfancy may have resulted in survivors developing severe long-term sequelae.


References

American Academy of Pediatrics. Committee on infectious diseases and committee on pe-

diatric AIDS. Measles immunization in HIV infected children. Pediatrics 1999; 103:

1057–1060.

Atkins MC, Esmonde TF. Guillain–Barre syndrome associated with rubella. Postgrad Med J

1991; 67(786): 375–376.

Banatvala JE. Health of mother, fetus and neonate following maternal infections during

pregnancy. In: Infections and Pregnancy (Coid CR, editor). London: Academic Press;

1977; pp. 437–488.

Banatvala JE, Best JM. Rubella. In: Principles and Practice of Clinical Virology (Zuckerman

AJ, Banatvala JE, Pattison JR, editors). 4th edition. Chichester: Wiley; 2000.

Banatvala JE, Brown DWG. The Lancet 2004; 363: 1127–1137.

Best JM, Banatvala JE. Personal Communication.

Best JM, Banatvala JE. Rubella. In: Principles and Practice of Clinical Virology (Zuckerman

AJ, Banatvala JE, Pattison JR, editors). 4th edition. Chichester: Wiley; 2000.

Best JM, Banatvala JE, Moore BM. Growth of rubella virus in human embryonic organ

cultures. J Hyg (Camb) 1968; 61: 407–413.

Boger III WP. Late ocular complications in congenital rubella syndrome. Ophthalmology

1980; 87: 1244.

Boger III WP, Petersen RA, Robb RM. Keratoconus and acute hydrops in mentally re-

tarded patients with congenital rubella syndrome. Am J Ophthalmol 1981; 91: 231.

Bosma TJ, Etherington J, O’ Shea S, et al. Rubella virus and chronic joint disease: is there an

association. J Clin Microbiol 1998; 36: 3524–3526.

Brown AS, Cohen P, Greenwald S, Sussex E. Nonaffective psychosis after prenatal exposure

to rubella. Am J Psychiatry 2000; 157: 3.

Buimovici-Klein E, Lang PH, Ziring PR, et al. Impaired cell mediated immune response in

patients with congenital rubella; correlation with gestational age at time of infection.

Pediatrics 1979; 64: 620–626.

Chang YC, Huang CC, Liu CC. Frequency of linear hyperechogenicity over the basal

ganglia in young infants with congenital rubella syndrome. Clin Infect Dis 1996; 22(3):

569–571.

Chantler JK, Smyrnis L, Tai G. Selective infection of astrocytes in human glial cell cultures

by rubella virus. Lab Invest 1995; 72: 334–340.

Chantler JK, Tingle AJ, Petty RE. Persistent rubella virus infection associated with chronic

arthritis in children. N Engl J Med 1985; 313: 1117–1123.

Clarke WL, Shaver KA, Bright GM, et al. Autoimmunity in congenital rubella syndrome. J

Pediatr 1984; 104: 370.

Cooper LZ. The history and medical consequences of rubella. Rev Infect Dis 1985; 7(Suppl

1): S1.

Craddock-Watson JE, Ridehalgh MKS, Anderson MJ, et al. Fetal infection resulting from

maternal rubella after the first trimester of pregnancy. J Hyg 1980; 85: 381–391.

Cremer NE, Oshiro LS, Weil MK, et al. Isolation of rubella virus from brain in chronic

progressive panencephalitis. J Gen Virol 1975; 29: 143–153.

Date M, Gondoh M, Katow S, Fukishima M, Nakamoto N, Kobayashi M, Abe T. A case of

rubella encephalitis: rubella virus genome was detected in the cerebrospinal fluid by po-

lymerase chain reaction. No To Hattatsu 1995; 27: 286–290.

J.E. Banatvala34

Deutman AF, Grizzard WS. Rubella retinopathy and subretinal neovascularization Am J

Ophthalmol 1978; 85: 82.

Eckstein M, Vijayalakshmi P, Killedar M, Gilbert C, Foster A. Aetiology of childhood

cataract in south India. Br J Ophthalmol 1996; 80(7): 628–632.

Enders G, Nickerl-Pacher U, Miller E, et al. Outcome of confirmed periconeptional maternal

rubella. Lancet 1988; I: 1445–1446.

Floret D, Rosenberg D, Hage GN, Monnet P. Hyperthyroidism, diabetes mellitus and the

congenital rubella syndrome. Acta Paediatr Scan 1980; 69: 259–261.

Forrest JM, Turnbull FM, Sholler GF, et al. Gregg’s congenital rubella patients 60 years

later. Med J Aust 2002; 177: 664–667.

Fortuin NJ, Morrow AG, Roberts WC. Late vascular manifestations of the rubella syn-

drome; a roentgenographic-pathologic study. Am J Med 1971; 51: 134.

Fraser JR, Cunningham AL, Hayes K, Leach R, Lunt R. Rubella arthritis in adults. Iso-

lation of virus, cytology and other aspects of the synovial reaction. Clin Exp Rheumatol

1983; 1(4): 287–293.

Frey TK. Neurological aspects of rubella virus infection. Intervirology 1997; 40: 167–175.

Geiger R, Fink FM, Solder B, Sailer M, Enders G. Persistent rubella infection after er-

roneous vaccination in an immunocompromised patient with acute lymphoblastic le-

ukemia in remission. J Med Virol 1995; 47(4): 442–444.

Ginsberg-Fellner F, Witt ME, Yahihashi S, et al. Congenital rubella syndrome as a model

for type 1 (insulin-dependent) diabetes mellitus: increased prevalence of islet cell surface

antibodies. Diabetologia 1984; 27(Suppl): 87–89.

Givens T, Lee DA, Jones T, Ilstrup DM. Congenital rubella syndrome: ophthalmic man-

ifestations and associated systemic disorders. Br J Ophthalmol 1993; 77: 358–363.

Gregg NM. Congenital cataract following German measles in the mother. Trans Op-

hthalmol Soc Aust 1941; 3: 35–46.

Harcourt GC, Best JM, Banatvala JE, Kennedy LA. HLA antigens and responses to rubella

vaccination. J Hyg (Camb) 1979; 83: 412–504.

Hastreiter AR, Joorabchi B, Pujatti G, et al. Cardiovascular lesions associated with con-

genital rubella. J Pediatr. 1967; 71(1): 59–65.

Heggie AD. Pathogenesis of the rubella exanthem: distribution of rubella virus in the skin

during rubella with and without rash. J Infect Dis 1978; 137: 74.

Karounos DG, Wolinsky JS, Thomas JW. Monoclonal antibody to rubella virus capsid

protein recognises a B cell antigen. J immunol 1993; 150(7): 3080–3085.

Lawn JE, Ref S, Baffoe-Bonnie B, Adadevoh S, Caul EO, Griffin GE. Unseen blindness,

unheard deafness, and unrecorded death and disability: congenital rubella in Kumasi,

Ghana. Am J Public Health 2000; 90(10): 1555–1561.

Lim KO, Beal DM, Harvey RL, Myers T, Lane B, Sullivan EV, Faustman WO.

Pfefferbaum. A. Brain dysmorphology in adults with congenital rubella plus schizo-

phrenic symptoms. Biol Psychiatry 1995; 37: 764–776.

Lobach DF, Hensley LL, Ho W, Haynes BF. Human cell antigen expression during the early

stages of fetal thymic maturation. J Immunol 1985; 135: 1752–1759.

Mauracher CA, Mitchel LA, Tingle AJ. Selective tolerance to the E1 protein of rubella virus

in congenital rubella syndrome. J Immunol 1993; 151: 2041–2049.

McIntosh ED, Menser MA. A fifty-year follow-up of congenital rubella. Lancet 1992;

340(8816): 414–415.

Mehl AL, Thomson V. Newborn hearing screening: the great omission. Pediatrics 1998; 101: 1–6.


Menser MA, Forrest JM, Bransbry RD. Rubella infection and diabetes mellitus. Lancet

1978; I: 57–60.

Menser MA, Harley JD, Hertzberg R, et al. Persistence of virus in the lens for 3 years after

prenatal rubella. Lancet 1967a; II: 387–388.

Menser MA, Reye RDK. The pathology of congenital rubella; a review written by request.

Pathology 1974; 6: 215.

Menser MA, Robertson SEJ, Dorman DC, et al. Renal lesions in congenital rubella. Pe-

diatrics 1967b; 40: 901.

Miller E, Cradock-Watson JE, Pollock TM. Consequences of confirmed maternal rubella at

successive stages of pregnancy. Lancet 1982; II: 781–784.

Minor PD, Muir P. Enteroviruses. In: Principles and Practice of Clinical Virology (Zucker-

man AJ, Banatvala JE, Pattison JR, Griffiths PD, Schoub B, editors). 5th edition. Chapter

4; 2004; 468–489.

Moriarty BJ. Childhood blindness in Jamaica. Br J Ophthalmol 1988; 72(1): 65–67.

Naeye RL, Blanc W. Pathogenesis of congenital rubella. JAMA 1965; 194: 1277–1283.

Oberfield SE, Cassulo AM, Chiriboga-Klein S, et al. Growth hormone dynamics in con-

genital rubella syndrome. Brain Dysfunc 1988; 1: 303.

O’Shea S, Best JM, Banatvala JE. A lymphocyte transformation assay for the diagnosis of

congenital rubella. J Virol. Methods 1992; 37: 139–148.

Ou D, Jonsen LA, Metzger DL, Tingle AJ. CD4 and CD8 T-cell clones from congenital

rubella syndrome patients with IDDM recognise overlapping GAD65 protein epitopes.

Hum Immunol 1999; 60: 652–664.

Preece MA, Kearney PJ, Marshall WC. Growth hormone deficiency in congenital rubella.

Lancet 1977; 2: 842.

Rand ML, Wright JF. Virus associated idiopathic thrombocytepenic purpura. Transfus Sci

1998; 19(3): 253–259.

Rawls WE, Desmyter J, Melnick JK. Virus carrier cells and virus free cells in fetal rubella.


Rawls WE, Melnick JL. Rubella virus carrier cultures derived from congenitally infected

infants. J Exp Med 1966; 123: 795–816.

Scasso CA, Bruschini L, Berrettini S, Bruschini P. Progressive sensorineural hearing loss

from infectious agents. Acta Ororhinolaryngol Ital 1998; 4(Suppl 59): 51–54.

Sever JL, South MA, Shaver KA. Delayed manifestations of congenital rubella. Rev Infect

Dis 1985; 7(Suppl 1): S164.

Smith A. Deafness and Hearing Impairment in Congenital Rubella Syndrome (CRS) Pre-

sented at the 5th Meeting of the WHO Steering Committee on Epidemiology and Field

Research, Geneva; 4–5 May 1999.

Smith CA, Petty RE, Tingle AJ. Rubella virus and arthritis. Rheum Dis Clin North Am

1987; 13: 265–274.

Steen E, Torp KH. Encephalitis and thrombocytopenic purpura after rubella. Arch Dis

Child 1956; 31: 470–473.

Tokugawa K, Ueda K, Fukushige J, Koyanagi T, Hisanaga S. Congenital rubella syndrome

and physical growth: a 17-year prospective, longitudinal follow-up in the Ryukyu Islands.

Rev Infect Dis 1986; 8(6): 874–883.

Tondury G, Smith DW. Fetal rubella pathology. J Pediatr 1966; 68: 867–879.

J.E. Banatvala36

Ueda K, Shingaki Y, Sato T, Tokugawa K, Sasaki H. Hemolytic anemia following post-

natally acquired rubella during the 1975–1977 rubella epidemic in Japan. Clin Pediatr

1985; 24: 155–157.

Van Straaten HL, Groote ME, Oudesluys-Murphy AM. Evaluation of an automated au-

ditory brainstem response infant hearing screening method in at risk neonates. Eur J

Pediatr 1996; 155(8): 702–705.

Webster WS. Teratogen update: congenital rubella. Teratology 1998; 58(1): 13–23.

Weil MJ, Itabashi J, Creamer NE. Chronic progressive panencephalitis due to rubella virus

simulating subacute scherosing panencephalitis. N Engl J Med 1975; 292: 994–998.

Wolinsky JS. Rubella. In: Fields Virology (Fields BN, Poknipe DM, Howley PM, Chanock

RM, Melinck J, Monath TP, Roizman B, Straus SE, editors). 3rd edition. Philadelphia:

Lippincott-Raven; 1996; pp. 899–929.

Woods WA, Johnson RT, Hostetler DD, Lepow ML, Robbins FC. Immunofluorescent

studies on rubella infected tissue cultures and human tissues. J Immunol 1966; 96:

253–260.

Ziring PR, Gallo G, Finegold M, et al. Chronic lymphocytic thyroiditis: identification of

rubella virus antigen in the thyroid of a child with congenital rubella. J Pediatr 1977; 90:

419–420.


Rubella Viruses



DOI 10.1016/S0168-7069(06)15003-X

39

Chapter 3

Laboratory Diagnosis of Rubella andCongenital Rubella

Jennifer M Besta, Gisela EndersbaKing’s College London School of Medicine at Guy’s, King’s College and St Thomas’Hospitals. St Thomas’ Hospital, London, SEI 7EH, UKbInstitute of Virology, Infectiology and Epidemiology e.V and Labor Prof. Dr. med.Gisela Enders and Partner, RosenbergstraX e 85, 70193 Stuttgart, Germany

Introduction

The accurate laboratory diagnosis of rubella infection in pregnant women presentingwith a rubella-like illness, or exposed to possible infection is essential, given that theclinical diagnosis of rubella is unreliable (Chapter 2). Rapid techniques are requiredto confirm infection in pregnancy so that a woman can be offered a termination ofpregnancy (TOP) or prenatal diagnosis (PD) as quickly as possible.

Rubella virus (RV) was first isolated in cell culture in 1962. Although RV growsin many cells, it does not produce a distinct cytopathic effect (CPE) in most cellcultures. Parkman et al. (1962) first isolated RV in primary vervet monkey kidney(VMK) cells, identifying it by interference with an enterovirus (Echovirus 11). In thesame year, Weller and Neva (1962) reported isolation of RV in primary humanamnion cells in which a CPE was produced. A distinct CPE is also produced in rabbitkidney cells (RK13) and these cells have been the most useful cell line for identi-fication of RV (McCarthy et al., 1963; Best and Banatvala, 1967). Rubella antibodieswere first measured by neutralization in VMK cells (Parkman et al., 1962). However,this test was too time consuming and labour intensive for routine use. The devel-opment of a haemagglutination inhibition (HAI) test in 1967 was a major advance,since this test was suitable for use in most laboratories and large numbers of seracould be tested within a short time (Halonen et al., 1967; Stewart et al., 1967).

Methods for the detection of rubella IgM were first developed for the rapiddiagnosis of rubella in the late 1960s. Initially sera were fractionated on sucrose

dx.doi.org/10.1016/S0168-7069(06)15003-X.3d

gradients or by gel filtration and the IgM-containing fractions tested for rubellaantibodies by HAI (Vesikari and Vaheri, 1968; Best and Banatvala, 1969). Thesetechniques were somewhat cumbersome and were replaced by radioimmunoassay(Tedder et al., 1982) and later by enzyme immunoassay (EIA) (Hodgson andMorgan-Capner, 1984; Enders and Knotek, 1986; Gerna et al., 1987), which aremore rapid and have higher throughput. Detection of rubella-specific IgM remainsthe method of choice for diagnosis of both postnatally and congenitally acquiredrubella, but isolation of virus and detection of viral RNA by nested reverse tran-scription-polymerase chain reaction (RT-nPCR) are useful for the diagnosis ofcongenital rubella in the fetus and newborn infant.

Laboratory techniques

Serological methods

There is little antigenic variation among RV strains circulating worldwide, thus,antigen prepared for serological tests with one strain will detect antibodies inducedby any RV strain.

Preparation of rubella antigens for use in serological assays

Currently RV antigens used for diagnostic assays are based on whole virus pro-duced in cell cultures. Continuous cell lines, such as BHK-21 and Vero, are bestsuited for conventional antigen production because of the higher levels of RVproduced (Best and O’Shea, 1995). However, even in these cell cultures RV doesnot grow consistently to high titres, making it difficult to obtain reproduciblebatches of antigen. This may be overcome by employing recombinant antigens,RV-like particles (RVLPs) and synthetic peptides: definition of optimal antigenshas focussed on a region of the immunodominant RV glycoprotein E1, whichcontains important B-cell epitopes responsible for inducing neutralizing and HAIantibodies (Wolinsky et al., 1993; Starkey et al., 1995). Recombinant antigensproduced in E. coli (Starkey et al., 1995), via baculovirus–insect cell expressionsystems (Schmidt et al., 1996; Nedeljkovic et al., 1999), viral-like particles synthe-sized in transfected cells (Hobman et al., 1994; Grangeot-Keros et al., 1995;Grangeot-Keros and Enders, 1997) and synthetic peptides representing the entireE1 or parts of it (Zrein et al., 1993; Pustowoit and Liebert, 1998; Cordoba et al.,2000; GieXauf et al., 2004) have been assessed for use in EIAs and immunoblotting.

RVLPs expressed in transfected BHK-21 cells carry epitopes for T-cell prolifer-ation and cytotoxicity, as well as B-cell epitopes, as they contain the three RV struc-tural proteins E1, E2 and C. These have been used in a commercial EIA for detectionof rubella-specific IgG and IgM. Studies comparing the IgG-EIA with the HAI andtwo whole virus-EIAs for specific IgG revealed a sensitivity of 94–100% and aspecificity of 80–98% (Grangeot-Keros et al., 1995). The sensitivity of the RVLPrubella IgM assay was 100% and the specificity 99.3%, when compared with two

J.M. Best, G. Enders40

other commercial EIAs employing whole virus antigen (Grangeot-Keros and Enders,1997).

Routine tests

Tests for rubella antibodies and specific IgG

Numerous techniques have been developed for the detection of rubella antibodies(Best and O’Shea, 1995). Their use depends on their practicality and turnaroundtime. Apart from neutralization, all the tests described below, will provide a resultwithin 24 h, but the appearance of rubella IgG in relation to clinical features willvary according to the test employed (see Fig. 1, p. 20 in Chapter 2).

The haemagglutination inhibition test, which is labour intensive, detects totalantibodies and its values are expressed in titres. The lipoproteins in serum areinhibitors of haemagglutination and must be removed from virus preparations usedas antigen and from test sera (Halonen et al., 1967; Stewart et al., 1967). Low false-positive results may be obtained if lipoprotein inhibitors are not completely re-moved from test sera. In the UK and USA, the HAI test is no longer used forrubella antibody screening and diagnosis, although it is still used in Germany andsome other European countries.

The enzyme immunoassay is the most common assay used for detection ofspecific IgG, as it is readily automated and commercial kits are available. The EIAtest kit of choice should have a high specificity (>98%) and a negative predictivevalue (NPV) of 100% (Pustowoit et al., 1996; Grangeot-Keros and Enders, 1997;L. Hesketh, personal communication, 2003).

The latex agglutination test (LA), which is available commercially, has theadvantage of rapidity as it gives qualitative results within a few minutes. Its

days post onset of rash

sera

0

10

20

30

40

50

60

70

80

90

100

1 2-3 4-15 16-28 29-42 43-70 71-120 121-180 181-272

% r

ub

ella

IgM

po

siti

ve

Rubazyme M

Enzygnost IgM

34 50 130 84 43 65 56 47 43

Fig. 1 Detection of rubella IgM following naturally acquired rubella in two commercial tests (552 sera

from 238 patients; Enders and Knotek, 1986).

Laboratory Diagnosis of Rubella and Congenital Rubella 41

sensitivity compared to the HAI test is 94–98% and it is therefore well suited forscreening purposes (Vaananen et al., 1985; Enders, 1987).

Single radial haemolysis (SRH) is less widely used now as it cannot be auto-mated and is subject to observer error. SRH plates are usually prepared in thelaboratory using commercially available reagents (Kurtz et al., 1980; Best andO’Shea, 1995). In Germany, the SRH test is commercially available in kit form.Results are available within 24 h. It may be used as a supplementary test for rubellaantibody screening to confirm low positive results. In addition, a negative value inthe SRH test (diameter o5mm) in combination with a low positive HAI result maybe indicative of primary acute infection.

Neutralization tests (NT), although labour intensive, are the reference standard fordetermining ‘‘protective’’ antibodies. Neutralizing antibodies are measured in somereference laboratories to verify immunity, especially when specific IgG levels are low(o10 IU/ml) following vaccination. The early neutralizing assays were carried out inprimary VMK cell cultures employing the interference technique (Parkman et al., 1964;Gitmans et al., 1983). Continuous cell lines such as RK-13 and Vero have also beenused, since RV causes subtle CPE and plaque formation in these cells. Tischer andGerike (2000) showed that the sensitivity of the plaque NT on RK-13 cells was nearlyequal to that of EIA because equivocal antibody concentrations in the EIA (4–6 IU/ml)could all be confirmed by plaque neutralization. In these conventional NT, results readby CPE or plaque reduction are available after 9–10 days.

A new NT has been described using Vero cells in microtitre plates. RV-positivecells are stained with monoclonal antibodies (against RV proteins E1, E2, C; ViralAntigens Inc., Tennessee, USA) and horseradish peroxidase-conjugated rabbit anti-mouse antibody, and results are available within 48h (M. Eggers, G. Enders et al.,unpublished data, 2004). Best and O’Shea (1995) have also described a microtitreplate assay using RK13 cells. Most of these NT-assays, which measure the capacity ofa given serum to block infection of cells in culture, are technically difficult and timeconsuming. Other approaches include EIAs based on the synthetic peptides SP15(Wolinsky et al., 1993; Pustowoit and Liebert, 1998; Cordoba et al., 2000) and BCH-178 (GieXauf et al., 2004). When HAI and SP15-EIA were compared for determiningimmune status, results demonstrated that SP15-EIA is very specific (100%) andsensitive (98%) for detecting protective antibodies (Cordoba et al., 2000). The BCH-178 EIA detects only a proportion of neutralizing antibodies, as the peptide hasinsufficient domains for binding neutralizing antibodies (GieXauf et al., 2004).

Detection of rubella IgM antibodies

Detection of rubella IgM is the method of choice for diagnosis of both postnatallyacquired and congenitally acquired rubella. Rubella IgM is detected by commercialEIA kits. ‘‘Indirect’’ and ‘‘IgM-capture’’ formats are available. The ‘‘indirect’’ formatemploys antigen adsorbed to the solid phase and enzyme-labelled anti-human IgM.This type of assay must employ a reagent to remove all rubella-specific IgG, in orderto avoid false-positive reactions with sera, which contain rheumatoid factor or otherIgM antiglobulins (Meurman et al., 1977; Vejtorp et al., 1979; Enders, 1985). The


‘‘IgM-capture’’ format employs anti-human IgM attached to the solid phase. Thus,IgG cannot bind and is washed away so that it does not contribute to false-positivereactions. Most IgM-capture assays employ native or recombinant rubella antigenand an enzyme-labelled monoclonal antibody to the antigen (Tedder et al., 1982;Gerna et al., 1987; Grangeot-Keros and Enders, 1997).

Information about development and persistence of specific IgM in the assayemployed, together with clinical details, are essential for interpretation of results,including the decision as to whether further sera are required. The sensitivity andspecificity of rubella IgM kits must be thoroughly evaluated using a panel ofpositive and negative sera, sera from patients with other infections and sera-con-taining autoantibodies (Matter et al., 1994; Hudson and Morgan-Capner, 1996).The performance of seven commercial assays was recently compared using well-defined panels comprising 283 sera from cases of rubella and 417 sera from non-rubella cases (Tipples et al., 2004). Sensitivity of both indirect and IgM captureassays, ranged from 74.1% to 78.9% and specificity from 85.6% to 97.2%. In serataken o10 days after onset of rash the sensitivity of the seven assays ranged from40% to 57.5% (mean 49.8%) and in convalescent phase sera taken Z10 days afteronset of rash from 94.3% to 98.8% (mean 97.2%). These results confirmed theexperience of earlier studies (Enders et al., 1985; Enders and Knotek, 1986), whichinclude some commercial assays still on the market. They confirmed that all testsperform well at the time of high rubella IgM concentration, but their sensitivitydiffers when very low rubella IgM concentrations are present, such as shortly afterthe onset of rash and when concentrations decline several weeks later (Fig. 1).

False-negative IgM results are more likely to occur in indirect assays and maybe due to high levels of specific IgG antibodies. The frequency of false-positiveresults will increase with decreasing positive predictive value when infection be-comes rare, e.g. when vaccination campaigns have been effective (Cutts andBrown, 1995). False-positive results may be due to

� cross-reacting IgM antibodies in sera from patients with measles, parvovirusB19, Epstein–Barr virus (EBV) and cytomegalovirus infections;

� polyclonal stimulation through EBV infection;� rheumatoid factor or other autoantibodies; and� heat inactivation of sera used in indirect assays.

Supplementary tests

Rubella IgG avidity

Indications for use. Tests for rubella IgG avidity are useful in the following sit-uations:

� To distinguish primary rubella in early pregnancy from rubella reinfection.


� In cases where rubella IgM has been detected in the absence of a rubella-likeillness or contact with a rubella-like illness, to distinguish primary rubella in-fection from ‘‘long-persisting rubella IgM antibodies’’(p. 45).

Principles. The avidity or functional affinity of rubella-specific antibodies in-creases with time after primary exposure to the antigen. Thus, specific IgG isinitially of low avidity, but will mature to high avidity within weeks or months afterinfection depending on the technique used.Methods to measure IgG avidity are based on EIA. The most common method

employs elution of low-avidity antibody from preformed immune complexes of IgGby addition of a denaturating agent to the washing buffer (elution principle—wash

method). Thomas and Morgan-Capner (1988) have used a different method, addingthe denaturating agent to the serum dilution buffer, to prevent the binding of low-avidity antibody to the antigen (dilution principle). Several denaturants such as urea(Hedman and Seppala, 1988; Enders and Knotek, 1989; Hedman and Rousseau,1989; Eggers et al., 2005) and diethylamine (DEA) (Thomas and Morgan-Capner,1988, 1991; Thomas et al., 1992a,b; Bottiger and Jensen, 1997; Enders, 2005) havebeen used for this purpose. Results of avidity tests are usually expressed as percentratio (Enders and Knotek, 1989; Bottiger and Jensen, 1997; Eggers et al., 2005) oras ratio of EIA absorbance values of a single serum dilution with and withoutdenaturation ( ¼ index calculation method) (Pustowoit and Liebert, 1998). Whenusing the DEA shift value method (DSV) (Thomas and Morgan-Capner, 1988,1991) results are expressed as distance between two curves drawn from absorbancesof serial sera dilutions with and without the protein denaturant.

Urea wash method. The urea wash method (elution principle) has been used tomeasure avidity after primary infection (Hedman and Seppala, 1988; Endersand Knotek, 1989; Hedman and Rousseau, 1989), rubella vaccination (Enders andKnotek, 1989) and rubella reinfection (Enders and Knotek, 1989; Hedmanand Rousseau, 1989). An enzyme-linked fluorescent assay (ELFA) using 6M ureawill soon be commercially available (VIDAS Rubella Avidity). A recent evaluationhas shown that an avidity index greater than 40% is indicative of high avidity andallows the exclusion of recent natural infection or vaccination o3–5 months beforeserum collection (Eggers et al., 2005).

DEA shift value method. This method (dilution principle) has been used exten-sively by Thomas and Morgan-Capner (Thomas and Morgan-Capner, 1988, 1991;Thomas et al., 1992a,b), who have also compared available methods to defineoptimal avidity testing conditions for differentiation between primary rubella, earlypostvaccination, natural or vaccine infection in the remote past, reinfection and‘‘persistent rubella-specific IgM reactivity’’ (Thomas et al., 1992a,b).

DEA wash method. The DEA wash method (elution principle) of Bottiger andJensen (1997) employs 35mM DEA and can be used with the Enzygnost anti-rubella-IgG-EIA (Dade Behring). It is important that patients’ sera are prediluted


to give a rubella IgG concentration of 70 IU/ml, because the rise of the avidityindex may be detected earlier and at incorrectly high values if there is a high specificIgG concentration (G. Enders, unpublished results).The kinetics of avidity maturation following primary infection and vaccination of

seronegative women are shown in Fig. 2. In patients with primary rubella there was arise from low (o30%) to moderate (30–50%) avidity 6–12 weeks after onset of rashand progression to high avidity (>50%) after about 4–5 months. Between 6 and 12months after infection 90–100% of patients had high-avidity antibody, which hasbeen shown to persist for many years. Avidity rises more slowly after rubella vac-

cination with high-avidity antibody detected in o10% of females 5 months aftervaccination, in 20–40% at 5–9 months and in 50% at 10–12 months. In approx-imately 30% of vaccinees avidity will remain at moderate levels for many years.When differentiating reinfection from primary rubella in pregnancy with the DEA

wash method, maturation of avidity is faster following reinfection (2–4 weeks afterrubella contact) than following a primary infection (about 4–5 months followingprimary rubella).The urea wash method has been extensively used, but comparative studies have

shown the DSV and DEA wash methods will detect low-avidity antibody for longerafter primary infection than the urea wash method (Table 1), thus prolonging thetime available for diagnosis of naturally acquired rubella. As the DEA washmethod is easier to perform than the DSV method, it is now the method of choice.

IgG avidity and long-persisting rubella-specific IgM. Rubella IgM antibodies maypersist for as long as 6 years and may be difficult to distinguish from IgM producedfollowing primary rubella and rubella reinfection (Enders, 1991; Thomas et al.,1992b; Bottiger and Jensen, 1997; Enders, 1997, 2005). If sera are obtained withinthe first 12–16 weeks of pregnancy, then avidity tests are useful to distinguish IgMantibodies due to primary rubella from long-persisting IgM antibodies. In those withlong-persisting IgM, the avidity indices will be moderate (mainly in vaccinees) orhigh (mainly post natural infection) irrespective of HAI and rubella IgG antibodylevels. This contrasts with primary infections where the rise from moderate-to-highavidity occurs 4–5 months after onset of infection. However, confirmation of long-persisting IgM comes from testing consecutive serum samples for 8 months or moreafter their initial detection.By testing more than 448 mother–infant pairs, it was shown that correctly defined

long-persisting rubella IgM antibodies in the mother were not a risk for congenitalinfection (Enders, 2005). Rubella IgM antibodies were found at the same concen-trations in the mothers at birth and during repeated testing in pregnancy, whilerubella IgM was not detected in newborn sera, thereby excluding congenital infection.

Non-reducing immunoblot (IB)

Non-reducing IB assays were developed for detection of rubella-specific immuno-globulins directed to topographical epitopes on the RV structural proteins (Zhang


et al., 1992). With a commercial non-reducing IgG IB (recomBlot Rubella IgG,MIKROGEN GmbH, Martinsried, Germany) based on recombinant rubella an-tigens (Hobman et al., 1994), it is possible to visualize IgG antibody reaction to theRV polypeptides C, E1, E2 and the dimer E1E2 by specific bands (Fig. 3). Duringthe first few days after onset of rash, nearly all sera develop IgG antibodies to Cand E1, but the E2 response is delayed (Meitsch et al., 1997). E2 antibodies appearapproximately 3–4 months after primary infection and approximately 5 monthsafter vaccination. In contrast to recent primary infection where antibodies to the E2

(A)

(B)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

<8 w

eeks

>8-1

0 wee

ks

>2,5

-3 m

onth

s

>3-4

mon

ths

>4-5

mon

ths

>5-6

mon

ths

>6-9

mon

ths

>9-1

2 m

onth

s

>1-2

year

s

>2-1

0 ye

ars

low avidity moderate avidity high avidity

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

<2 w

eeks

>2-4

wee

ks

>4-6

wee

ks

>6-8

wee

ks

>2-3

mon

ths

>3-4

mon

ths

>4-6

mon

ths

>6-9

mon

ths

>9-1

2 m

onth

s

low avidity moderate avidity high avidity

Fig. 2 Distribution of rubella-specific IgG avidity at various time points in (A) patients with primary

rubella infection (203 sera from 105 patients) and in (B) previously seronegative rubella vaccinees (278

sera from 159 vaccinees, most of whom were vaccinated with RA27/3) (modified DEA wash method; G.

Enders, unpublished results, 2004).


band develop in 90% (5–6 months after infection), o60% of vaccinees developantibodies to E2 even 2–10 years after vaccination (Fig. 4).

The development of IgG antibodies against the E2 protein has proved a usefuladditional assay for differentiating reinfection from primary infection, since in

Table 1

Percentage of sera with low-avidity rubella IgG1; comparison of three methods using 206 sera from 171

patients with primary rubella (data from Fig. 1 of Thomas et al., 1992a)

Method Time after onset of acute rubella infection

r14

days

15–28

days

Z1–2

months

Z2–3

months

Z3–4

months

Z4–5

months

Z5–7

months

>7

months

DSV (%) 100 97 94 78 91 71 21 0

DEA

wash (%)

�98 �95 �75 �44 �38 0

Urea

wash (%)

�80 �62 �24 �10 0

serum control

time postrubella exanthem: 10 days 57 days >5 months

Dimer E1 E2

structural protein E1

structural protein E2

capsid protein C

6 6 2

Fig. 3 Kinetics of the IgG response to the rubella polypeptides C, E1 and E2 in consecutive sera from a

patient with rubella infection, visualized by specific bands in a commercial non-reducing IgG immuno-

blot (recomBlot Rubella, MIKROGEN).


reinfection the E2 band is present within about 4 weeks after contact, in contrast to3 months after primary infection. Non-specific binding may occur in about 8% ofsera (G. Enders, unpublished results). In patients with cow’s milk allergy, anti-bodies to bovine proteins may react with milk protein used in the IB-buffers.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

<8 w

eeks

>8-1

0 wee

ks

>2,5

-3 m

onth

s

>3-4

mon

ths

>4-5

mon

ths

>5-6

mon

ths

>6-9

mon

ths

>9-1

2 m

onth

s

>1-2

year

s

>2-1

0 yea

rs

E2 band detectable

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

<2 w

eeks

>2-4

wee

ks

>4-6

wee

ks

>6-8

wee

ks

>2-3

mon

ths

>3-4

mon

ths

>4-6

mon

ths

>6-9

mon

ths

-12

mon

ths

E2 band detectable

(A)

(B)

Fig. 4 Immunoblot analysis of the IgG response to recombinant E2 protein in (A) patients with primary

acute rubella infection (255 sera of 129 patients) and in (B) previously seronegative rubella vaccinees (269

sera of 168 vaccinees, of whom most were vaccinated with RA27/3) (non-reducing IgG immunoblot:

recomBlot Rubella, MIKROGEN; G. Enders, unpublished results, 2004).


Alternative specimens for diagnosis and surveillance

Oral fluid. Oral fluid consists of secretions from the salivary glands and crevicularfluid. It can be used for detection of virus-specific antibodies, since IgG and IgMantibodies reflect those in the serum (Mortimer and Parry, 1991; McKie et al.,2002) as well as for detection of virus. Oral fluid can be easily collected from themouth using an absorbent device (Fig. 5); several such devices are available com-mercially (Vyse et al., 2001).Since the collection of oral fluid is a non-invasive technique, it is particularly

useful for children. In the UK, oral fluid tests are used to investigate rash illness inpopulations where measles and rubella are rare, due to MMR vaccination pro-grammes. In a recent UK study, none of the 195 children (aged under 16 years)with a rash illness had measles or rubella IgM, and other infections, includingparvovirus B19, group A streptococcus and human herpesvirus-6 were confirmed in48% of these children (Ramsay et al., 2002).

Fig. 5 Oral fluid collection devices. From top to bottom: Orasure, Omni-SAL, Oracol (Reprinted from

Vyse et al. 2001 with permission from the author and the Royal Institute of Public Health).


Oral fluid is particularly useful for seroepidemiological studies, as it is easier tocollect and a phlebotomist and sterile equipment including needles are not required(Mortimer and Parry, 1991; Nokes et al., 1998). In developing countries whereneedles and syringes are often reused, this reduces the risk of transmitting blood-borne infections. Methods for the extraction, transportation and preservation ofsamples have been established (Mortimer and Parry, 1991; Nokes et al., 1998;Morris et al., 2002).Sensitive assays are required to detect specific antibodies in oral fluid, as an-

tibody concentrations are much lower than in serum. Consequently antibody cap-ture assays are preferred for detection of specific IgG and IgM. Rubella-specificIgG or IgM antibody capture radioimmunoassays (GACRIA or MACRIA) wereused initially, but the IgG assay has since been adapted to amplification-based EIA(Vyse et al., 1999). It is possible to use a modified commercial assay with antigen onthe solid phase to measure rubella-specific IgG in oral fluid (Ben Saleh et al., 2003).Detection of rubella-specific IgG and IgM antibodies in oral fluid compare wellwith serum (Perry et al., 1993; Ramsay et al., 1998; de Oliveira et al., 2000). Oralfluid tests are more efficient at detecting antibodies in children than in adults(Nokes et al., 1998); they are not sufficiently sensitive and specific for rubellasusceptibility screening of adults. In order to detect rubella IgM, oral fluid shouldbe collected between 1 and 5 weeks after onset of rash. Oral fluid can also be usedfor the diagnosis of congenital rubella (Eckstein et al., 1996), and the measurementof IgG avidity (Akingbade et al., 2003). The widespread use of oral fluid tests forrubella awaits the development of suitable commercial assays, such as those re-cently developed for measles and mumps.Using RT-nested PCR (RT-nPCR; p. 52) it is possible to detect RV RNA in oral

fluid collected with the Oracol device (Fig. 5; Malvern Medical Developments,Worcester, UK), for which oral fluid should be collected within 7 days after onsetof rash (Jin et al., 2002). This method is also useful for the diagnosis of congenitalrubella and can be used to obtain RV RNA for sequencing for molecular epide-miological purposes (Jin et al., 2002; Vyse and Jin, 2002; Cooray et al., 2006).

Dried blood spots (DBS). DBS have also been used successfully for detection ofrubella IgM and IgG antibodies (Helfand et al., 2001; Karapanagiotidis et al.,2005) and viral RNA (Pitcovski et al., 1999; De Swart et al., 2001). DBS arecollected by finger prick onto high-quality filter paper (e.g. Schleicher and Schuell,903). Guthrie cards can also be used. They can be stored at room temperature(Parker and Cubitt, 1999). Packing and transportation of DBS is easier than sera,as they are not considered as ‘‘Dangerous Goods’’.

Virological methods

Detection of RV is seldom used for diagnosing postnatal rubella, since serologicalmethods are quick and reliable. RT-nPCR is preferred for PD and detection of RVin congenital rubella cases due to its greater sensitivity and rapidity (see below).


However, virus isolation from clinical specimens is helpful to obtain sufficient RVfor sequencing, e.g. in order to carry out phylogenetic analysis, and to differentiatewild-type and vaccine virus (Bosma et al., 1996; Frey et al., 1998; Zheng et al., 2003;World Health Organisation, 2005; Cooray et al., 2006).

Virus isolation

This procedure is time-consuming and labour-intensive and only carried out inspecialized laboratories. Results may not be available for up to 4 weeks. Specimensshould have two passages in Vero cells, with subsequent identification of RV inRK13 cells, in which a distinct CPE is usually produced (Fig. 6).

Alternatively RV may be identified in the cell cultures by indirect immunoflu-orescence using hyperimmune or monoclonal antibodies (Best and O’Shea, 1995).Punctate staining throughout the cytoplasm occurs with antibodies to the C proteinand intense staining of the Golgi apparatus with antibodies to the E2 and E1proteins. Alternatively RV-positive cells can be identified by RT-nPCR (Revelloet al., 1997). Vero/SLAM cells can be used for the isolation of both measles virusand RV. Thus, it has been suggested that measles virus-negative cell cultures, couldbe used for the detection of RV by indirect immunofluorescence (Best et al., 2005).

Recently a novel cell-culture-based assay for RV (genotypes I and II) detectionin clinical specimens was described, which combines cell culture techniques, the useof genetically modified RV genes and reporter gene expression (Tzeng et al., 2005).Preliminary evaluation demonstrated good sensitivity and specificity, but thismethod may not be suitable for routine use.

Fig. 6 Cytopathic effect caused by rubella virus in RK13 cells 3 days after inoculation.


Reverse transcription-polymerase chain reaction

The high GC content of the RV genome has made the development of a reliableRT-PCR difficult and a variety of protocols were validated to find the most suitablefor different purposes. Diagnostic applications include invasive PD and diagnosisof congenital rubella (pp. 58, 66). Highly sensitive PCR assays, detecting 3–10copies of RV RNA, are required for PD as fetal specimens generally contain lowRV copy numbers. The RT-nPCR protocol described by Bosma et al. (1995a,b) hasbeen used fairly widely. This RT-nPCR amplifies a 143 bp sequence from a well-conserved region of the E1 gene and has a sensitivity of r2 RNA copies. OtherRT-nPCR protocols targeting the E1 gene have been used successfully for PD(Revello et al., 1997; Mace et al., 2004). Revello et al. (1997) compared direct RT-nPCR, culture-RT-nPCR and virus isolation and found sensitivities of 100%, 75%and 75%, respectively. Studies have also shown that RT-nPCR is about 20% moresensitive than virus isolation for PD. A high reproducibility of RT-nPCR resultswith fetal specimens stored for 1–4 years at �701C has been observed using the RT-nPCR of Bosma et al. (1995a,b).

Specimens should be transported at 41C or on dry ice or liquid nitrogen if intransit for >12 h, as viral RNA is easily destroyed at ambient temperatures. Toavoid false-negative results, several strategies, including spiking and non-compet-itive internal controls, have been established to monitor PCR inhibitors. For PD itis especially important to check the quality of nucleic acid extraction, because thePCR efficiency can be dramatically reduced by the presence of PCR-inhibitorysubstances (e.g. haem or added anticoagulants such as heparin). Non-competitiveinternal controls such as keratin mRNA (Bosma et al., 1995a) are widely used,however, these endogenous sequences may vary from one sample to another andunderestimate the amount of PCR inhibitor in the sample. For this reason a com-petitive PCR, in which a synthetic construct is used, is generally favoured. This isamplified by the same primers, but detected by a different molecular probe and isable to monitor both the quality of nucleic acid extraction and PCR (Revello et al.,1997).

A longer amplicon is required for genotyping of RV isolates for molecularepidemiological studies. A RT-nPCR, which amplifies a 592-bp sequence from theE1 gene has been described for this purpose (Cooray et al., 2006). A multiplexnPCR to detect measles virus, RV and parvovirus B19 with the detection limit of 10genome equivalents may be useful in countries where these viruses are circulating(del Mar Mosquera et al., 2002). Viral RNA is stable on certain filter papers forlater investigation by RT-PCR, which is useful for specimens sent to distant lab-oratories.

Facilities for rubella testing and their distribution

In countries, such as UK, USA, Finland and Sweden, where there is currently littleor no rubella circulation and limited demand for diagnosis, most hospital/local


laboratories provide tests for rubella antibody screening (including confirmatoryassays) and rubella IgM. Alternative techniques for rubella IgM, IgG and IgGavidity assays, IB, virus isolation and tests for PD should be available in referencelaboratories, which should also be responsible for evaluating available assays. InGermany, France and North Italy such additional tests are usually performed bynational reference and university laboratories as well as by private expert labo-ratories. Specialized tests are required for use with oral fluid.

A global measles–rubella laboratory network has been established by WHO,which comprises global specialized laboratories, regional reference laboratories,national and subnational laboratories (Robertson et al., 2003). Laboratory sup-port, training and quality control are available for this network. In addition toconducting serological assays, these laboratories are encouraged to obtain RVisolates for molecular epidemiological studies (Best et al., 2005).

Rubella antibody screening

In the UK, rubella antibody screening is offered to all pregnant women and towomen with no history of rubella or MMR immunization or a previous positiverubella antibody test. In Germany, rubella antenatal screening is obligatory in earlypregnancy since 1972. The aim of the antenatal screening programmes is to identifywomen who are susceptible to rubella, for whom vaccine is advised postdeliveryand not to identify or exclude rubella infection in the current pregnancy. It isparticularly important that screening is offered to women who come from countrieswhere they may not have been offered rubella vaccination (Tookey et al., 2002;Department of Health, 2003; Robert-Koch-Institut, 2004). In the UK, testing isconsidered unnecessary if there is documented previous evidence of two reliabletests on different serum samples both confirming the presence of rubella-specificIgG (Department of Health, 2003). However, most UK clinics may find it moreconvenient to test all antenatal patients regardless of past history of testing orvaccination. In Germany, antenatal antibody testing is considered unnecessary ifthere is documented HAI titre of Z1 : 32 or an IgG antibody level Z15 IU/mlbefore the current pregnancy. Apart from that, all women should be tested prior toin vitro fertilization, irrespective of previous test results or vaccination history.

In the UK, many other European countries and the USA, the main tests usedfor antibody screening are commercially available EIAs and LA and SRH may beused as second-line/confirmatory assays. In contrast, in Germany the HAI testassisted by IgG EIAs is for the time being the main screening test. The cut-off forrubella antibody detection is currently 10 IU/ml (Skendzel, 1996; Department ofHealth, 2003), or an HAI titre of 1 : 32 in Germany. Sera giving low levels (o10 IU/ml, or HAI titre 1 : 8/1 : 16) or equivocal results on initial testing should be re-testedusing an alternative assay which may also help to monitor the reliability of the firstassay (O’Shea et al., 1999). In the UK, screening results are reported as ‘‘rubellaantibody detected/not detected’’ rather than ‘‘immune/susceptible’’ (Department ofHealth, 2003). The presence of rubella-specific IgG does not exclude recent


infection and if a woman has a history of rash or contact with a rash further testsare required.

If a low level (o10 IU/ml, or HAI titre 1 : 8/1 : 16) of rubella antibody isdetected by two reliable assays in a woman with a documented history of two ormore rubella vaccinations, further doses of vaccine are unlikely to be of value aspersistent rises in antibody concentration are seldom achieved (Matter et al., 1997;Tischer and Gerike, 2000), and protection against rubella and congenital rubella isassumed. Nevertheless such women are advised to report any rash illness or contactwith a rubella-like rash, so that further investigations can be carried out (Morgan-Capner and Crowcroft, 2002).

Diagnosis of rubella

Postnatally acquired rubella

Serological investigation for postnatally acquired rubella, especially in pregnancy,is essential to confirm or refute rubella infection. Measles, enteroviruses, HHV-6and HHV-7, group A streptococcus and dengue virus may present with a similarrash, and parvovirus B19, chikungunya-, Ross River-, West Nile- and Sindbisviruses may present with both rash and joint symptoms (Banatvala and Brown,2004).

Rubella is usually diagnosed serologically. Detection of rubella IgM in a serumtaken 3–6 days after onset of rash is the method of choice for diagnosis of acuterubella. Although specific IgM may be detected earlier in some patients, this willdepend on the assay used (p. 43). If no specific IgM is detected in a serum takeno6days after onset, it is advised to collect a second serum a few days later. This shouldbe tested in parallel with the first serum for specific IgM and IgG, as a significantrise in rubella IgG concentration may be detected in addition to rubella IgM.Rubella IgG antibodies are usually detectable by well-controlled commercial EIAswithin 7 days after onset of rash while HAI antibodies may be detected within 2days (see Chapter 2, Fig. 1, p. 20).

If rubella-specific IgM is detected in a patient with rash, this is likely to indicaterecent primary rubella, but care is needed in interpretation (pp. 42–43). Thepatient’s history and results of any previous tests should be considered and sup-plementary tests, previously described, may be required to confirm the diagnosis(Best et al., 2002).

Diagnosis of women with rubella-like illness in pregnancy

The risk of congenital rubella syndrome (CRS) is greatest in the first 12 weeks ofpregnancy (Chapter 2). Therefore, it is particularly important to test women whoare in the first trimester, rapidly, so that if they wish to consider a pregnancytermination following a diagnosis of rubella, this can be done as early as possible.Close collaboration between clinical and laboratory staff is essential if serological


tests are to be interpreted accurately. In order to interpret laboratory resultsaccurately, the following information is required:

� date of last menstrual period (LMP);� previous documented history of rubella antibody screening and rubella/MMR

vaccination;� date and duration of contact;� type of contact (e.g. significant: household or at work);� age of index case; and� symptoms or rash and rubella/MMR vaccination history in the index case.

Serum should be collected as soon as possible after onset of symptoms and testedfor rubella-specific IgG and IgM antibodies. A second serum is usually required5–10 days later to demonstrate seroconversion and to confirm the presence of IgM,if this was detected in the first sample. Accurate information on the date of onset ofillness and the distribution of rash, lymphadenopathy and arthropathy (Chapter 2),should be obtained in order to interpret serological results (see Fig. 1, p. 20).

The results on sera from women who present more than 4–6 weeks after onsetof symptoms may be difficult to interpret, as specific IgM may have becomeundetectable in about 20% of cases depending in part on the sensitivity of the IgMassay used. The diagnosis in such cases may be clarified by applying the IgG avidityand IB tests to sera obtained in the first 12–16 weeks of gestation. In sera taken inthis time period, moderate-to-high IgG avidity together with the E2 band on IBconfirms that IgG and IgM antibodies have not been acquired by recent primaryinfection in the current pregnancy (Enders, 2005).

Assessment of women exposed to rubella-like illnesses in pregnancy

UK Public Health Laboratory Service joint working party of the advisory com-mittees of virology and vaccines and immunization has produced guidelines on themanagement of rash illness in pregnancy (Morgan-Capner and Crowcroft, 2002).Women who have been exposed to a rubella-like illness are more likely to acquireinfection if exposure is close and prolonged, such as within their own household orat work. Thus, in populations without successful childhood vaccination pro-grammes, children may acquire infection at school and present a risk to a sero-negative mother. Women who have had only a brief exposure may usually bereassured initially, although testing is still recommended.

A documented history of two previous positive IgG antibody tests or two doc-umented doses of rubella vaccine or one documented dose of vaccine followed by apositive IgG antibody result, indicates past rubella infection or vaccination (Bestet al., 1989). Although further rubella testing seems unnecessary, in practice a womanwith significant exposure would be tested. Seronegative women and/or those withlow levels of antibody (o10 IU/ml) should be retested at 7–10 day intervals for up to4 weeks after contact. When the pregnant woman is seronegative, it may be desirable


to collect serum or oral fluid from the index case, in order to confirm or refute thediagnosis of rubella. If the index case is found to be seronegative further follow-up ofthe pregnant woman is unnecessary, but she should be vaccinated postpartum.

It is always useful to test stored serum samples from the current or a previouspregnancy, as such a serum can be tested in parallel with the later serum sample(Morgan-Capner and Crowcroft, 2002). This is possible in the UK and some othercountries where it is recommended to store the antenatal screening sera at �201Cfor at least 1 year (Department of Health, 2003). If there is no documented in-formation and also no stored serum sample available, the advice given in thealgorithm (Fig. 7) should be followed. However, it may be easier and quicker totake advantage of the early use of supplementary tests.

Diagnosis of rubella reinfection in pregnancy

Rubella reinfection is defined as a rubella-specific immune response in someonewith previous naturally acquired or vaccine-induced rubella. Reinfection is usuallysubclinical and occurs mainly in persons with vaccine-induced immunity who areclosely exposed to rubella. Maternal rubella reinfection and its possible risk of fetalinfection and damage was an important diagnostic issue in such countries as theUK, USA, Germany and Israel in the late 1980s and 1990s when rubella continuedto circulate among children (Bullens et al., 2000). Since then the risk of exposure of

verify low IgG values withan alternative test andcollect a further serum within 10 days

3-4 days after onset ofacute infection?

Differentiate:• acute infection• acute reinfection• longpersisting IgM• non-specific IgM

IgG titer rise and IgMdetected:

Consider reinfection andadvise accordingly → askurgently for vaccine history

rubella IgG detected:IgG>10 IU/ml (HAI >1:32)rubella IgM not detected

IgG detectedIgM detected

IgG not detectedIgM continues positive, sometimes until delivery

Non-specific IgM !

rubella IgG detected at low level IgG <10 IU/ml (HAI 1:8/1:16)rubella IgM not detected

previous infection orvaccination

no further testing necessary

Non-specific IgM ?If test values are unchanged:

send a third serum at17th/ 18th week of gestationto exclude titer rise due toreinfection following previousvaccination

Test for rubella-specific IgM and IgG

IgG seroconversionIgM titer rise

primary acute rubella

IgG not detectedIgM detected

Obtain further serum within 10days

advise on risk and prenatal diagnosis if necessary

Fig. 7 Algorithm for diagnosis of pregnant women exposed to rubella-like rash.


pregnant women to circulating RV has decreased due to increased vaccinationcoverage in these countries. The risk of congenital rubella defects after subclinicalmaternal reinfection in the first 12 weeks of pregnancy has not been preciselydetermined. It has been estimated to be o10% and probably o5% (Morgan-Capner et al., 1991; Morgan-Capner and Crowcroft, 2002). Congenital rubelladefects have not been reported after reinfection occurring beyond 12 weeks’ ges-tation. Thus, it is important to distinguish rubella reinfection from primary infec-tion in the first 12 weeks of pregnancy, as the risk of fetal damage after reinfectionis considerably lower than after primary infection (>80%). Invasive PD is useful toassess whether fetal infection has occurred. Although symptomatic maternal re-infection is very rare, it is considered to pose a higher risk to the fetus compared toasymptomatic maternal reinfection (Morgan-Capner and Crowcroft, 2002), as ob-served in symptomatic and asymptomatic primary rubella infection (Miller et al.,1982).

Maternal reinfection is diagnosed by a significant rise in rubella IgG and/orHAI concentration, sometimes to very high levels, in a woman with pre-existingantibodies (Best et al., 1989; Enders and Knotek, 1989; Morgan-Capner et al.,1991). Rubella IgM is usually detected if sera are collected within 4–6 weeks ofcontact. Ideally, pre-existing antibodies should be confirmed in a stored serumsample. However, if such a serum is not available, evidence of pre-existing antibodymay be accepted if there are at least two previous laboratory reports of antibodiesZ10 IU/ml obtained by reliable techniques, or a single serum with rubella anti-bodies Z10 IU/ml obtained after documented rubella vaccination (Best et al.,1989). Supplementary tests like IgG avidity and immunoblotting may be used todistinguish reinfection from primary infection (pp. 44–47). It is less safe to excludeprimary rubella in early pregnancy if serum is obtained beyond the 16th weekof gestation. Characteristic antibody responses seen in reinfection are shown inFig. 8.

Diagnosis of congenitally acquired rubella

Prenatal diagnosis

The value of invasive PD is limited by the fact that virological and serologicalinvestigations have to be delayed until relatively late in gestation (21–24 weeks),and for 7–8 weeks after the onset of the rubella-like illness. Furthermore, thenumber of studies in which sensitive techniques have been employed to detectcongenital rubella infection (Table 2) are relatively small and few have reported theoutcome of pregnancy, which is necessary for calculating sensitivity, specificity,positive and negative predictive values of available techniques. Thus, it is difficultto compare the reliability of investigations employing different specimens from theconceptus. Additionally, the duration of pregnancy at which TOP may be carriedout varies from country to country.


No official guidelines exist for applying invasive PD to the diagnosis of con-genital rubella infection. However, these techniques have been in use for 15 yearsand are of value in the following cases:

� Laboratory confirmed primary rubella in the early second trimester (12–18weeks gestation).

� When primary rubella in the first trimester cannot be excluded due to equivocalrubella IgM results, despite the use of supplementary tests (pp. 44–47).

� Laboratory confirmed rubella reinfection before 12 weeks’ gestation.

Invasive PD is NO longer required in the following situations:

� Laboratory confirmed primary rubella before the 12th week of gestation, sincethe high risk of fetal infection and damage is well established (Chapter 2) and themajority of parents are disinclined to continue with the pregnancy.

� When a seronegative pregnant woman has been inadvertently vaccinated with arubella-containing vaccine, as this has not been associated with symptomaticCRS (Chapter 4).

� Long-persisting IgM antibodies are detected in pregnant women, since they arenot associated with congenital infection (pp. 45–46).

day post contactWG

13_14

20_15

36_18

86_25

154_34

190delivery

6_13

IgG avidity

highpositive

positive

lowpositive

negative

C

E1

E2

IgG avidityIgGHAINTIgM

high

moderate

low

IgG immunoblot

maternal sera

-+-

++

(+)

+++

+++

++

(+)

(+)+-

-+-

+++

contact symptoms

newbornCord blood

Prenatal diagnosis: RV RT-nPCR neg. in amniotic fluid

IgG

IgM

HAI

NT

Fig. 8 Schematic antibody pattern for routine and supplementary serologic assays in confirmed rubella

reinfection. Results from a case of confirmed symptomatic rubella reinfection in the 13th week of

pregnancy following previous vaccination at age 19 and exposure to own child with clinical and se-

rologically confirmed rubella. The baby had no evidence of congenital infection and was healthy at birth

and at age 5. IgG immunoblot: +, -antibody is detectable; (+) possibly detectable; –, not detectable;

WG, week of gestation.


Table 2

Results of prenatal diagnosis after acute rubella infection in early pregnancy by RT-nPCR: confirmed/not confirmed by outcome of pregnancy

Study Pregnant women RV RNA detected RV RNA not detected

na With

symptom

na Confirmedb Not

confirmedbLost to

follow-up

na Confirmedb Not

confirmedbLost to

follow-up

Bosma et al.

(1995b), UK

7 4 1 1 2 3 — 1 2

Tanemura et al.

(1996), Japan

34 21 8 5 — 3 26 20 4 2

Revello et al.

(1997), Italy

(retrospective)

8 8 8 6 — 2 0 — — —

Katow (1998),

Japan

253 112 48 22 4 22 205 130 13 62

Henquell et al.

(1999), France

6 6 1 1 — — 5 2 — 3

Mace et al.

(2004), France

45 20 20 11 — 9 25 6 1 18

Enders et al.

(1994–2004),

Germany

77 77 58 33 1 24 19 8 1 10

aNumber of cases tested.bCriteria of congenital infection: termination of pregnancy: RV RNA and/or RV detection in fetal tissue; live-born infant: CRS and specific IgM antibody

detection at birth; no CRS-like symptoms and specific IgM antibody; detection at birth; and persistence of specific IgG antibodies until 10 months of age.

La

bo

rato

ryD

iag

no

siso

fR

ub

ellaa

nd

Co

ng

enita

lR

ub

ella59

Invasive PD is usually accompanied by ultrasound evaluation. However, in the caseof maternal rubella, diagnosis of fetal infection on the basis of ultrasound findingshas not been reported, even though certain fetal abnormalities (e.g. cerebral vent-riculomegaly with intracranial calcifications, cardiac malformations, microcephaly)could potentially be detected by ultrasound (Crino, 1999; Enders et al., unpub-lished data, 2006). The failure to detect abnormalities by ultrasound is mainly dueto the fact that pregnancies with acute rubella up to the 12th week of gestation aremainly terminated before the above-mentioned ultrasound abnormalities becomevisible beyond the 20th week of gestation.

There seems to be little point in conducting invasive PD if TOP is unacceptableto the parents. However, in such cases, a negative PD result will provide somereassurance to the mother during the inevitably anxious time. Before performing PDit is essential that the risks of the sampling techniques as well as the limited value ofpositive PD results for prognosis of damage in the newborn are explained to themother by those responsible for her care. It should be clarified that in case of positivePD results, the risk of fetal damage depends mainly on the gestational week ofmaternal infection.

Fetal samples used for prenatal diagnosis

Chorionic villus biopsies, amniotic fluid and fetal blood (Table 3) have been usedfor detection of RV RNA by RT-nPCR, which can provide a result within 48 hours(p. 52). Fetal blood can also be used for the detection of rubella IgM (p. 42).

� Amniotic fluid is considered the optimal sample for PD as amniocentesis is lessinvasive than fetal blood sampling and amniotic fluid can be obtained in largeramounts than fetal blood. Ideally, amniotic fluid samples should be taken 7–8weeks after rubella infection and after the 21st week of gestation (Enders et al.,2001; Mace et al., 2004).

� Chorionic villus biopsies have not been used as extensively as amniotic fluid asdetection of RV may not reflect fetal infection as infection of the villi is likely tooccur before infection of fetal tissues. A positive result may also be due to RV incontaminating maternal blood or decidual tissue. However, this can be clarified byDNA typing with short tandem repeats for differentiation of maternal and fetaltissue or blood. The optimal time for collecting chorionic villus biopsies is notknown, but collection should not be performed before 11th week of gestation onaccount of the risk of limb-reduction-defects (Jauniaux and Rodeck, 1995). RVRNA positive results have been obtained 2–10 weeks after onset of maternal rash.When a negative result is obtained from chorionic villi, amniotic fluid or fetalblood should be obtained later in pregnancy to confirm this result.

� Fetal blood, taken by cordocentesis, can be used for detection of RV by RT-nPCR as well as for detection of rubella IgM. It should be collected 7–8 weeksafter onset of maternal rubella and not before the 21st week of gestation.


Although IgM antibodies may be produced by the fetus as early as the 12thweek of gestation, rubella IgM levels will be too low to be detected before 21weeks (Enders and Jonatha, 1987; Jacquemard et al., 1995; Enders, 2005). He-parin should not be used when collecting fetal blood samples, as it may inhibitthe PCR reaction.

All specimens should be tested for RV RNA by RT-nPCR in triplicate at least, andtwo or three assays should ideally be used for IgM testing. Observing these rec-ommendations will reduce false-negative RT-nPCR and rubella IgM results. False-negative RT-nPCR results may be due to the fact that virus may not be present infetal samples at detectable concentrations at the time of sampling.

Table 3

Fetal samples for the prenatal diagnosis of congenital rubella infection

Sample Optimal time

for sampling

(weeks of

gestation)

Detection

of RV RNA

by RT-

nPCRa

Technical

risk of fetal

loss

Comments

Sensitivity Specificity

Amniotic

fluid

Z21 86% 91% 0.2–1% Sampling possible

after 15 weeks,

but virus more

likely to be

detected at 21

weeks or more

(31/36)b (10/11)c

Chorionic

villus

biopsy

11–16 83%

(5/6)bInsufficient

data

0.2–3% Positive result

may be due to

contamination

with maternal

blood or decidual

tissue

Fetal blood Z21 73% 100% 0.5–2% Collection is

technically more

demanding. Can

also be used for

detection of

rubella IgM (see

Table 4)

(16/22)b (7/7)c

aUnpublished results from G. Enders and colleagues.bNumber of true positives/total number of fetuses or newborns congenital infected.cNumber of true negatives/total number of fetuses or newborns congenital uninfected.


Practical considerations

When results are available, the virologist should discuss them with the obstetricianand parents, in order to help the mother decide whether to continue with thepregnancy. From the few published studies of PD, there is insufficient informationavailable on the reliability of RT-nPCR and rubella IgM detection in fetal samples.Results should therefore be interpreted with caution. Ultrasound beyond the 22ndweek of gestation may be helpful in excluding major defects, but it cannot be usedfor affirming fetal infection.

Seven studies of invasive PD of congenital rubella are listed in Table 2. Henquellet al. (1999) and Mace et al. (2004) tested only amniotic fluid and fetal blood, but inthe other five studies amniotic fluid, chorionic villus biopsies and fetal blood sampleswere included. In the study by Enders et al. results were also assessed by pregnancyoutcome (Table 4). In this study, a case was designated as congenital rubella infectionif one fetal sample (chorionic villus biopsy, amniotic fluid or fetal blood) was RVRNA positive by RT-nPCR, or rubella IgM was detected in fetal blood. Ninety-seven

Table 4

Prenatal diagnosis in cases with confirmed primary rubella infection: Detection rate for RV RNA using

RT-nPCR (E1 primers) in various fetal samples and/or for rubella-specific IgM in fetal blood and

calculation of sensitivity, specificity and predictive value based on the outcome of pregnancy (Enders et

al., unpublished results)

No. positive/total no. tested (%) Cases

RT-nPCRa Rubella IgMa

58–77 (75) 32–46 (70)

With known outcome (n) 43 29

Sensitivityb(%) 97 (33/34) 86 (19/22)

Specificityc(%) 89 (8/9) 100 (7/7)

Positive predictive value

(PPV)d(%) 97 (33/34) 100 (19/19)

Negative predictive value

(NPV)e(%) 89 (8/9) 70 (7/10)

Note: Criteria for congenital infection:

Termination of pregnancy: RV RNA and/or rubella virus detection in fetal tissue; live-born infant: CRS

and specific IgM antibody detection at birth; no CRS-like symptoms and specific; IgM antibody de-

tection at birth; and persistence of specific IgG antibodies to 7–11th months of age.aRT-nPCR was performed with the protocol of Bosma et al. (1995a) and samples were tested in du-

plicate. FB was tested for rubella-specific IgM antibodies using three different commercial rubella IgM

test kits, two based on the IgM-capture principle. IgM testing was done in serum dilution recommended

by the manufacturer and in a lower dilution mainly in duplicate.bSensitivity: (number of true positives/total number of fetuses or newborns congenital infected)� 100.cSpecificity: (number of true negatives/total number of fetuses or newborns uninfected)� 100.dPPV: (number of true positives/number of positive test results)� 100.eNPV: (number of true negatives/number of negative test results)� 100.


per cent of those with positive RT-nPCR results were associated with fetal infection,while 89% of those with negative RT-nPCR results were uninfected. False-positiveand false-negative RT-nPCR results have occurred in all studies (Table 2). Thus, anegative result in a single sample for PD, even if carried out under optimal conditions,can never exclude fetal infection.

Borderline results may be obtained when maternal infection has occurred be-tween 12 and 18 weeks of gestation, particularly if infection was subclinical, as viralload and rubella IgM concentrations are more likely to be low in these situations(G. Enders, personal communication). In these cases repeat testing and collectionof further samples are necessary, as demonstrated by Tang et al. (2003), whodescribed negative results in amniotic fluid taken at 19 and 23 weeks gestation, butRV RNA and rubella IgM detected in fetal blood taken at 23 weeks. However,collection of further samples will delay diagnosis and must always be weighedagainst the risks incurred when carrying out invasive prenatal tests (Table 3).

Neonatal diagnosis and diagnosis in later infancy

Neonates born to women who have had rubella at any stage of pregnancy should betested for evidence of congenital infection, as should neonates born with clinicalfeatures compatible with congenital rubella (Chapter 2). When these children areassessed age, clinical findings, maternal history, rubelliform illnesses since birth aswell as laboratory results should be taken into consideration. Results of laboratorytests are required to confirm congenital rubella (Miller et al., 1994; World HealthOrganisation, 1999; Reef et al., 2000). Rubella IgM, persistence of specific IgGbetween 7 and 11 months of age, and detection of virus excretion by virus isolation orRT-nPCR are established methods for the diagnosis of congenital rubella (Table 5).

When there is a history of rubella or unconfirmed rubella-like illness in themother, the neonate should be tested for rubella IgM using a sensitive IgM-captureassay, as this method appears to be more sensitive for the diagnosis of congenitalrubella than indirect IgM EIAs (Chantler et al., 1982; Corcoran and Hardie, 2005).Rubella IgM may also be detected in oral fluid (Eckstein et al., 1996). Infants bornwith signs or symptoms suggestive of rubella-like defects should also be tested atthe earliest opportunity, since there is a risk of RV transmission to susceptiblepersons.

Specific IgM is detected in most cases of CRS tested within the first 3 monthsafter birth with all types of IgM tests (Table 6). If a low or equivocal IgM result isobtained, a further specimen should be obtained within a week and tested in ad-ditional IgM-capture tests. A negative IgM results in the neonatal period and up to 3months of age will virtually exclude congenital infection. For samples collected be-tween 3 and 18 months of age, IgM-capture EIAs are the most sensitive and specificmethods available. Beyond 18 months rubella IgM is rarely detected. In asympto-

matic congenital infection rubella IgM is, in our experience, detected up to 3 monthsof age at rates similar to those of CRS, but more rarely beyond 3 months of age.


Laboratory confirmation of congenital rubella infection may also be obtained byisolation of RV or detection of RV RNA in nasopharyngeal swabs, urine and oralfluid. Although they were using relatively insensitive methods for virus isolation,Cooper and Krugman (1967) isolated RV from nasopharyngeal secretions of mostcases of congenital rubella up to the age of 1 month, from 62% at 1–4 months, 33%at 5–8 months, 11% at 9–12 months and only 3% at 13–20 months of age. In morerecent studies we have confirmed that RT-nPCR was more sensitive than virusisolation for all samples, with virus excretion detected more frequently and for longerin CRS cases (Table 7). The latest positive RT-nPCR result was obtained from aurine sample obtained at 23 months of age. In asymptomatic congenital rubellainfection RV RNA was detected by RT-nPCR less frequently and only up to 3months of age. RT-nPCR has also been successfully employed to detect RV RNA inpostmortem fetal tissues (Bosma et al., 1995b; Revello et al., 1997). Specimens can be

Table 5

Laboratory techniques for the diagnosis of congenital rubella infection

Established methods Prenatal

diagnosis

Diagnosis in

infancy

References

Chantler et al. (1982)a

Cooper (1975)a

Rubella-specific IgM

in serum

Yes Birth–1 year Thomas et al. (1993)a Enders

et al. (personal

communication)a

Rubella-specific IgM

in oral fluid

No Birth–1 year Eckstein et al. (1996)a

Rubella-specific IgG No 7 months–1

year

Cooper and Krugman

(1967)a

RV isolation in cell

culture

Yesb Birth–1 year

Enders et al. (personal

communication)a

Bosma et al. (1995b)a, c

Tanemura et al. (1996)c

Revello et al. (1997)c

RT-nPCR (RV RNA

detection)

Yes Birth–1 year Katow (1998)c

Mace et al. (2004)c

Enders et al. (personal

communication)a,c

aDiagnosis in infancy.bToo slow to be useful.cPrenatal diagnosis.


Table 6

Frequency and persistence of rubella-specific IgM in congenitally infected infants with clinical features of CRS:data from four studies

Reference Technique for

detection of rubella

IgM

No. sera positive/no. tested (%)

Age (months)a

o3 3–6 6–12 12–18 18–24 Total tested

Cooper (1975) SDG+HAI 36/60 (60%) 10/23 (43.5%) 12/31 (38%) 5/21 (23%) 1/21 (5%) 143

Chantler et al.

(1982)

MACRIA 35/35 (100%) 6/7 (85.7%) 13/21 (62%) 9/19 (47.4%) 1/17 (5.9%) 99

MACEIA 35/35 (100%) 4/7 (57%) 6/21 (28.6%) 2/19 (10.5%) 1/17 5.9%)

Thomas et al.

(1993)

MACRIA 51/51 (100%) 11/13 (84.6%) 7/23 (30.4%) 4/15 (26.7%) 1/24 (4%) 126

Enders et al.

(unpublished

data,

1985–2004)

Commercial IgM-

capture and

indirect IgM EIAs

91/95 (96%) 13/15 (87%) 10/14 (71%) 3/4 (75%) 4/6 (67%) 134

aAge ranges are approximate, as different age ranges were used in each study. SDG ¼ sucrose density gradient centrifugation; HAI ¼ haemagglutination

inhibition; MACRIA ¼ IgM-capture radioimmunoassay; and MACEIA ¼ IgM-capture enzyme immunoassay.

La

bo

rato

ryD

iag

no

siso

fR

ub

ellaa

nd

Co

ng

enita

lR

ub

ella65

transported from distant sites on dry ice or liquid nitrogen, and for RT-nPCR (butnot virus isolation) they can be dried on filter paper (De Swart et al., 2001).

When cataracts are removed the lens aspirates can be used for detection of RV.RT-nPCR has been successfully employed to detect RV RNA in lens aspirates frominfants aged 2–12 months with CRS (Bosma et al., 1995b). RV RNA detection hasnot been fully evaluated for detection of RV in older children with eye disease.

Prompt diagnosis of congenital rubella is more difficult when there is no historyof a rubella-like illness in pregnancy, especially if the infant presents at more than 3months of age, when it may no longer be possible to detect rubella IgM, isolate RVor detect RV RNA. This situation arises when children have isolated sensorineuraldeafness detected in later infancy. Such children should be tested for rubella IgG at7–11 months of age, when maternal IgG antibody will normally have disappearedand MMR vaccination has not yet been received (Fig. 9); the detection of specificIgG is suggestive of congenital infection. The value of testing sera from older childrenwill depend on the epidemiology of rubella locally. Although rubella is rare in chil-dren under the age of two in temperate climates, it is more frequent in tropical,developing countries. Congenital rubella infection may be excluded in most sero-negative infants and children although some congenital rubella patients will loserubella antibody (see below). Diagnosis of congenital rubella in children who havereceived MMR vaccination is unlikely to be possible, although failure to developrubella antibodies after vaccination is suggestive of congenital rubella (p. 68).

RV SP15 peptide EIA and rubella IB may also be useful for the diagnosis ofcongenital rubella and have been used to define serological differences between

Table 7

Rubella virus excretion detected by RT-nPCR and virus isolation in 22 cases of CRS (Enders et al.,

unpublished results, 1993–1996)

RT-nPCR Months after birth

0–3 4–6 7–12 13–24 Total

% (pos/n) % (pos/n) % (pos/n) % (pos/n) % (pos/n)

Sample

Nasopharyngeal secretions 84 (16/19) 100 (2/2) 0 (0/1) 50 (3/6) 75 (21/28)

Urine 76 (19/25) 58 (7/12) 43 (3/7) 60 (3/5) 65 (32/49)

EDTA-blood 46 (6/13) 44 (4/9) 0 (0/2) 0 (0/1) 40 (10/25)

Cerebrospinal fluid 33 (2/6) 25 (1/4) — 25 (1/4) 29 (4/14)

Virus isolation

Nasopharyngeal secretions 37 (7/19) 0 (0/2) 0 (0/1) 0 (0/6) 25 (7/28)

Urine 36 (9/25) 33 (4/12) 14 (1/7) 40 (2/5) 33 (16/49)

EDTA-blood 25 (1/4) 33 (1/3) — 0 (0/1) 25 (2/8)

Cerebrospinal fluid 25 (1/5) 50 (2/4) — 25 (1/4) 31 (4/13)

pos/n ¼ no. positive/no. tested.


infants with symptomatic and asymptomatic congenital rubella infection (Meitsch etal., 1997; Pustowoit and Liebert, 1998). In the 10 children with CRS born to motherswith infection in the first 12 weeks of gestation, antibodies to SP15 and the E2glycoprotein were detected less frequently and in significantly lower concentrationsthan in the 8 asymptomatic congenital infected newborns of mothers with infectionbeyond the 10th week of gestation. However, these assays need further evaluation.

Detection of low-avidity IgG1 by the DEA-shift method between 6 months and 3years of age may aid the diagnosis, as the avidity matures more slowly in childrenwith CRS than following postnatal infection (Thomas et al., 1993). Tests for cell-mediated immunity may also be useful. In a study by O’Shea et al. (1992) 10 of the 13(79.9%) children under the age of 3 with CRS did not respond in a rubella-specificlymphoproliferative assay, compared with only 1 of the 13 (7.6%) seropositive con-trol children. Although this assay is more demanding than EIA, it may be useful forthe diagnosis of congenital rubella in children under the age of 3. More recently, theenzyme-linked immunospot test (ELISPOT) has been used to detect rubella-specificcellular immunity (B. Pustowoit et al., personal communication, 2005). Unfortu-nately it is necessary to obtain heparinized blood for all cell functional assays and toperform the test within a very short time (24h) after blood sampling.

Loss of rubella antibodies in congenital rubella. The loss of rubella antibodies wasfirst reported by Cooper et al. (1971), who showed that 50 of the 270 (18.5%)

Maternal Infection

Maternal IgG

Birth

Infant IgM

Infant IgG

Infant lymphoproliferativeresponse

1st 2nd 3rd 2 4 6 1 2 3 4

Trimesters Months YearsAge of Child

Fig. 9 Serological markers for diagnosis of congenital rubella. (Reproduced from Diagnostic Proce-

dures for Viral, Rickettsial and Chlamydial Infections, 7th edition with permission from the American

Public Health Association.)


congenitally infected children had lost HAI antibodies by the age of 5. When 19 ofthese seronegative children were given HPV77 rubella vaccine, 17 (89%) failed toseroconvert. Since then Ueda et al. (1987) reported that 13 of the 381 children withcongenital rubella became seronegative by HAI between 3 and 17 years of age.Forrest et al. (2002) reported that 13 of 32 (41%) of Gregg’s original congenitallyinfected patients were seronegative by EIA at the age of 60. Enders and colleagueshave found that five hearing impaired pregnant women aged 21–34 years withcongenital rubella infection and low levels of HAI, IgG and NT antibodies, failedto respond to first or repeated vaccination and in three of the five women tested byIB, the E2 band—a tentative marker for immunity—was absent (G. Enders, per-sonal communication, 2003). Best and O’Shea (unpublished observations) alsonoted that adult women with a history of congenital rubella infection failed toseroconvert when vaccinated. Thus, the lack of rubella antibodies in older childrenmay not exclude congenital rubella, but failure to respond to vaccine, may provideevidence of congenital rubella infection in a patient with rubella-like defects.

Remaining diagnostic problems

Protective antibodies

Further research is required to identify the type of specific antibody associated withprotection from reinfection. This is of particular concern for populations withvaccine-induced immunity, which may wane in the absence of the booster effectsdue to contact with circulating RV. While previous natural rubella infection isgenerally considered to induce lifelong protection, this cannot be guaranteed fol-lowing vaccination, although it is probable that immunological memory exists.

The reason why reinfection occurs in some individuals is not clear, since intranasalchallenge studies have not identified any antibody type (NT, HAI, IgG, IgA) or titre,which correlates with protection against reinfection (Harcourt et al., 1980; O’Shea etal., 1983, 1994). Similarly, RV-specific lymphoproliferative responses do not appear tobe associated with protection (O’Shea et al., 1994). Studies of the E1 open readingframe of RV strains from cases of reinfection suggest that these strains do not differfrom other strains (Frey et al., 1998). The absence of antibodies against known linearepitopes or unknown conformational epitopes on the E1 and E2 glycoproteins areassumed to be associated with lack of protective immunity (Wolinsky et al., 1993;GieXauf et al., 2004). Studies of immune responses following rubella reinfection sug-gest that lack of E2 antibodies may be a marker of susceptibility to reinfection,however, more data are required to confirm this hypothesis.

Standardization

Standardization of most rubella methods is required. Commercial and ‘‘in-house’’HAI tests differ by the use of different erythrocytes, types of antigen and method ofabsorption of patients’ sera. Various rubella IgG EIAs are used worldwide for


rubella antibody screening, but differences in calibration have affected the inter-pretation of results and have made the definition of an IgG cut-off level difficult(Pustowoit et al., 1996; Matter et al., 1997). While European kit manufacturers usethe WHO second international standard for anti-rubella serum (Statens Serum-institut, Copenhagen) or national substandards, until 1997 most US manufacturerscalibrated their tests by the CDC standard (e.g. Gunter et al., 1996). As a result,some US tests achieved IgG concentrations that were twice as high as those ofseveral European test kits (Pustowoit et al., 1996; G. Enders, unpublished obser-vations). Thus, results from the same patient may have been reported as rubellaantibodies ‘‘detected’’ in a US and ‘‘not detected’’ in a European rubella IgG EIA.This may also happen in patients with a IgG antibody value around cut-off level atrepeat testing several months apart—even with the same test and in the samelaboratory. In such cases the question of immunity may be decided on a history ofat least one documented vaccination. In a study by the European Sero-Epidemi-ology Network involving seven laboratories, large differences in unitage were foundeven though all laboratories were using a standardized EIA method (Andrews etal., 2000), suggesting that other factors like antigen type and quality play a role inachieving uniform IgG results.

Diagnosis of congenital rubella

Detection of rubella IgM with IgM-capture techniques is usually a reliable tech-nique for the diagnosis of congenital rubella in the first 3 months of life (seeNeonatal diagnosis and diagnosis in later infancy). However, there is a demand forthe evaluation of new techniques for diagnosis of congenital rubella after 3 monthsin children with deafness only, as they may not present in the neonatal period, asdiscussed above.

Conclusions and outlook

As a result of WHO recommendations (World Health Organisation, 2000), pro-grammes for surveillance of rubella and congenital rubella have been established inmany countries, with a continuing demand for techniques for the diagnosis ofrubella and congenital rubella (World Health Organisation, 1999). Most problemsencountered with rubella diagnosis can now be solved with the improved andexpanded range of serological methods described in this chapter.

Although many countries have substantially reduced the incidence of rubellaand congenital rubella, unsubstantiated fears about the safety of MMR vaccine incountries such as the UK, have led to a fall in uptake of MMR to below 80%. Thisis not high enough to prevent circulation of RV and may result in some unvac-cinated girls remaining susceptible into their childbearing years. Recent rubellaoutbreaks have also been seen among two connected religious groups in the Neth-erlands and Canada, who refuse vaccination. Thus, it is important to maintain andimprove techniques for the laboratory diagnosis of rubella and congenital rubella.


In addition, antenatal rubella antibody screening should be continued, even incountries with well-established vaccination programmes, as it is of particular im-portance for immigrant women from countries with endemic rubella and no rubellavaccination or poorly implemented vaccination programmes. Recent reports ofsymptomatic congenitally infected infants born to such women in European coun-tries should draw attention to the fact that CRS is down but not out (Tookey et al.,2002; Tookey, 2004; G. Enders, personal observation, 2003–2005).

Acknowledgements

We wish to thank Simone Exler and Marion Biber of the Laboratory Professor G.Enders and Partners for their excellent assistance in preparing and correcting thischapter and Dr. Siobhan O’Shea, Guy’s and St Thomas’ NHS Foundation Trustfor review of the manuscript.

References

Akingbade D, Cohen BJ, Brown DWG. Detection of low-avidity immunoglobulin G in oral

fluid samples: new approach for rubella diagnosis and surveillance. Clin Diagn Lab

Immunol 2003; 10: 189–190.

Andrews N, Pebody RG, Berbers G, et al. The European sero-epidemiology network:

standardizing the enzyme immunoassay results for measles, mumps and rubella.

Epidemiol Infect 2000; 125(1): 127–141.

Banatvala JE, Brown DWG. Rubella. Lancet 2004; 363: 1127–1137.

Ben Saleh A, Zaatour A, Pomery L. Validation of a modified commercial assay for the

detection of rubella-specific IgG in oral fluid for use in population studies. J Virol Meth-

ods 2003; 114: 151–158.

Best JM, Banatvala JE. A comparison of RK13, vervet monkey kidney and patas monkey

kidney cell cultures for the isolation of rubella virus. J Hyg (Camb) 1967; 65: 263–271.

Best JM, Banatvala JE. Serum IgM and IgG responses in postnatally acquired rubella.

Lancet 1969; 1: 65–68.

Best JM, Banatvala JE, Morgan-Capner P, Miller E. Fetal infection after maternal rein-

fection with rubella: criteria for defining reinfection. Br Med J 1989; 299: 773–775.

Best JM, Castillo-Solorzano C, Spika JS, et al. Reducing the global burden of congenital

rubella syndrome. Report of the WHO Steering Committee on Research Related to Measles

and Rubella Vaccines and Vaccination, June 2004. J Infect Dis 2005; 192(11): 1890.

Best JM, O’Shea S. Rubella virus. In: Diagnostic Procedures for Viral, Rickettsial and

Chlamydial Infections (Lennette EH, Lennette DA, Lennette ET, editors). 7th ed. Wash-

ington, DC: American Public Health Association 1995; pp. 583–600.

Best JM, O’Shea S, Tipples G, et al. Interpretation of rubella serology in pregnancy—pitfalls

and problems. Br Med J 2002; 325: 47–48.

Bosma TJ, Best JM, Corbett KM, et al. Nucleotide sequence analysis of a major antigenic

domain of 22 rubella virus isolates. J Gen Virol 1996; 77: 2523–2530.


Bosma TJ, Corbett KM, Eckstein MB, et al. Use of PCR for prenatal and postnatal di-

agnosis of congenital rubella. J Clin Microbiol 1995b; 33: 2881–2887.

Bosma TJ, Corbett KM, O’Shea S, et al. PCR for detection of rubella virus RNA in clinical

samples. J Clin Microbiol 1995a; 33: 1075–1079.

Bottiger B, Jensen IP. Maturation of rubella IgG avidity over time after acute rubella

infection. Clin Diagn Virol 1997; 8(2): 105–111.

Bullens D, Smets K, Vanhaesebrouck P. Congenital rubella syndrome after maternal re-

infection. Clin Pediatr (Phila) 2000; 39(2): 113–116.

Chantler S, Evans CJ, Mortimer PP, Cradock-Watson JE, Ridehalgh MKS, Cooper LZ,

Florman AL, Ziring PR, Krugman S. A comparison of antibody capture radio- and

enzyme immunoassays with immunofluorescence for detecting IgM antibody in infants

with congenital rubella. J Virol Methods 1982; 4: 305–313.

Cooper LZ. Congenital rubella in the United States. In: Progress in Clinical and Bio-

logical Research (Krugman S, Gerhson A, editors). 3. New York: Alan R Liss, Inc; 1975;

pp. 1–21.

Cooper LZ, Forman AL, et al. Loss of rubella hemagglutination inhibition antibody in

congenital rubella. Failure of seronegative children with congenital rubella to respond to

HPV-77 rubella vaccine. Am J Dis Child 1971; 122: 397–403.

Cooper LZ, Krugman S. Clinical manifestations of postnatal and congenital rubella. Arch

Ophthalmol 1967; 77: 434.

Cooray S, Warrener L, Jin L. Improved RT-PCR for diagnosis and epidemiological sur-

veillance of rubella. J Clin Virol 2006; 35: 73–80.

Corcoran C, Hardie DR. Serologic diagnosis of congenital rubella: a cautionary tale. Pediatr

Infect Dis J 2005; 24(3): 286–287.

Cordoba P, Lanoel A, Grutadauria S, Zapata M. Evaluation of antibodies against a rubella

virus neutralizing domain for determination of immune status. Clin Diagn Lab Immunol

2000; 7(6): 964–966.

Crino JP. Ultrasound and fetal diagnosis of perinatal infection. Clin Obstet Gynecol 1999;

42: 71–80.

Cutts F, Brown DWG. The contribution of field tests to measles surveillance and control: a

review of available methods. Rev Med Virol 1995; 5: 35–40.

de Oliveira SA, Siqueira MM, Brown DWG. Diagnosis of rubella infection by detecting

specific immunoglobulin M antibodies in saliva samples: a clinic-based study in Niteroi

RJ, Brazil. Rev Soc Bras Med Trop 2000; 33: 335–339.

De Swart RL, Nur Y, Abdallah A. Combination of reverse transcriptase PCR analysis and

immunoglobulin M detection on filter paper blood samples allows diagnostic and epi-

demiological studies of measles. J Clin Microbiol 2001; 39(1): 270–273.

del Mar Mosquera M, de Ory F, Moreno M, Echevarria JE. Simultaneous detection of

measles virus, rubella virus, and parvovirus B19 by using multiplex PCR. J Clin Microbiol

2002; 40: 111–116.

Department of Health Screening for Infectious Diseases in Pregnancy. Standards to Support

the UK Antenatal Screening Programme. London: Stationery Office; 2003. Available

from http://www.dh.gov.uk/assetRoot/04/09/20/49/04092049.pdf.

Eckstein MB, Brown DWG, Foster A, et al. Congenital rubella in south India: diagnosis

using saliva from infants with cataract. Br Med J 1996; 312: 161.

Eggers M, Enders M, Strobel S, et al. A novel and rapid method for determination of rubella

virus immunoglobulin G avidity. Clin Microbiol Infect 2005; 11(Suppl 2): 33.


http://www.dh.gov.uk/assetRoot/04/09/20/49/04092049.pdf

Enders G. Problems of rubella diagnosis by various IgM techniques and the need for test

combinations. In: Rapid Methods and Automation in Microbiology and Immunology

(Habermehl KO, editor). Berlin: Springer; 1985; pp. 146–160.

Enders G. Vergleich eines einfachen Latex-Agglutinationstestes mit dem Roteln-Haagglu-

tinationshemmtest und Hamolysis-in-Gel-Test. Lab Med 1987; 11: 221–223.

Enders G. Diagnostik von Rotelninfektionen in der Schwangerschaft durch konventionelle,

immunologische und molekular-biologische Methoden. In: Neues in der Virusdiagnostik

(Deinhardt F, Maass G, Spiess H, editors). Richard-Haas-Symposium der DVV; 1991;

pp. 133–152.

Enders G. Qualitatssicherung in der Serodiagnostik bei der Mutterschaftsvorsorge: Qua-

litatssicherung und aktuelle Aspekte zur Serodiagnostik der Roteln in der Schwangersc-

haft. Symposium Moderne Aspekte der Mikrobiologischen Diagnostik, Kurzfassungen

von Vortragen des 3. Symposium am 04. Dezember 1996 in Berlin. Clin Lab 1997; 43:

1019–1032.

Enders, G. Mutterliche Infektionen mit dem Risiko der pra- und perinatalen Ubertragung—

Teil I. Labormedizinische Aspekte wichtiger Infektionen im Uberblick. Gynako-

logie+Geburtshilfe 06, 2005; 6: 38–46.

Enders G, Bader U, Lindemann L, et al. Prenatal diagnosis of congenital cytomegalovirus

infection in 189 pregnancies with known outcome. Prenat Diagn 2001; 5: 362–377.

Enders G, Jonatha W. Prenatal diagnosis of intrauterine rubella. Infection 1987; 15:

162–164.

Enders G, Knotek F. Detection of IgM antibodies against rubella virus: comparison of two

indirect ELISAs and an anti-IgM capture immunoassay. J Med Virol 1986; 19: 377–386.

Enders G, Knotek F. Rubella IgG total antibody avidity and IgG subclass-specific antibody

avidity assay and their role in the differentiation between primary rubella and rubella

reinfection. Infection 1989; 17(4): 218–226.

Enders G, Knotek F, Pacher U. Comparison of various serological methods and diagnostic

kits for the detection of acute, recent, and previous rubella infection, vaccination, and

congenital infections. J Med Virol 1985; 16: 219–232.

Forrest JM, Turnbull FM, Sholler GF, et al. Gregg’s congenital rubella patients 60 years

later. Med J Aust 2002; 177: 664–667.

Frey TK, Abernathy ES, Bosma TJ, et al. Molecular analysis of rubella virus epidemiology

across three continents, North America, Europe and Asia, 1961–1997. J Infect Dis 1998;

178: 642–650.

Gerna G, Zannino M, Revello M, et al. Development and evaluation of a capture enzyme-

linked immunosorbent assay for determination of rubella immunoglobulin M using

monoclonal antibodies. J Clin Microbiol 1987; 25: 1033–1038.

GieXauf A, Letschka T, Walder G, et al. A synthetic peptide ELISA for the screening of

rubella virus neutralizing antibodies in order to ascertain immunity. J Immunol Methods

2004; 287: 1–11.

Gitmans U, Enders G, Gluck H, et al. Rotelnschutzimpfung mit HPV77DE5- und RA 27/3-

Impfstoffen von 11–16 Jahre alten Madchen: Untersuchungen zur Antikorperentwicklung

und Antikorperpersistenz nach 4 und 8 Jahren. Immun Infekt 1983; 11: 79–90.

Grangeot-Keros L, Enders G. Evaluation of a new enzyme immunoassay based on recom-

binant rubella virus-like particles for detection of immunoglobulin M antibodies to

rubella virus. J Clin Microbiol 1997; 35(2): 398–401.


Grangeot-Keros L, Pustowoit B, Hobman T. Evaluation of Cobas Core Rubella IgG EIA

recomb, a new enzyme immunoassay based on recombinant rubella-like particles. J Clin

Microbiol 1995; 33: 2392–2394.

Gunter EW, Lewis BG, Koncikowski SM. Laboratory Procedures Used for the Third Na-

tional Health and Nutrition Examination Survey (NHANES III), 1988–1994, U.S. de-

partment of health and human services (http://www.cdc.gov/nchs/data/nhanes/nhanes3/

cdrom/nchs/manuals/labman.pdf); 1996.

Halonen PE, Ryan IH, Stewart JA. Rubella hemagglutinin prepared with alkaline extraction

of virus grown in suspension culture of BHK-21 cells. Proc Soc Exp Biol Med 1967; 125:

162–167.

Harcourt G, Best JM, Banatvala JE. Rubella specific serum and nasopharyngeal antibodies

in volunteers with naturally acquired and vaccine induced immunity following intranasal

challenge. J Infect Dis 1980; 142: 145–155.

Hedman K, Rousseau SA. Measurement of avidity of specific IgG for verification of recent

primary rubella. J Med Virol 1989; 27: 288–292.

Hedman K, Seppala I. Recent rubella virus infection indicated by a low avidity of specific

IgG. J Clin Immunol 1988; 8: 214–221.

Helfand RF, Keyserling HL, Williams I, Murray A, Mei J, Moscatiello C, Icenogle J, Bellini

WJ. Comparative detection of measles and rubella IgM and IgG derived from filter paper

blood and serum samples. J Med Virol 2001; 65(4): 751–757.

Henquell C, Bournazeau JA, Vanlieferinghen P, et al. Recrudescence en 1997 des infections

rubeoliques pendant la grossesse. La Presse Med 1999; 28(15): 777–780.

Hobman TC, Lundstrom ML, Mauracher CA, et al. Assembly of rubella virus structural

proteins into virus-like particles in transfected cells. Virology 1994; 202(2): 574–585.

Hodgson J, Morgan-Capner P. Evaluation of a commercial antibody capture enzyme

immunoassay for the detection of rubella specific IgM. J Clin Pathol 1984; 37: 573–577.

Hudson P, Morgan-Capner P. Evaluation of 15 commercial enzyme immunoassays for the

detection of rubella-specific IgM. Clin Diagn Virol 1996; 5: 21–26.

Jacquemard F, Hohlfeld P, Mirlesse V, et al. Prenatal diagnosis of fetal infections. In:

Infectious Diseases of the Fetus and Newborn Infant (Remington JS, Klein JO, editors).

Philadelphia: WB Saunders Company; 1995; pp. 99–107.

Jauniaux E, Rodeck C. Use, risks and complications of amniocentesis and chorionic villous

sampling for prenatal diagnosis in early pregnancy. Early Pregnancy 1995; 1(4): 245–252.

Jin L, Vyse A, Brown DW. The role of RT-PCR assay of oral fluid for diagnosis and

surveillance of measles, mumps and rubella. Bull World Health Organ 2002; 80: 76–77.

Karapanagiotidis T, Riddell M, Kelly H. Detection of rubella immunoglobulin M from

dried venous blood spots using a commercial enzyme immunoassay. Diagn Microbiol

Infect Dis 2005; 53: 107–111.

Katow S. Rubella virus genome diagnosis during pregnancy and mechanism of congenital

rubella. Intervirology 1998; 41: 163–169.

Kurtz JB, Mortimer PP, Mortimer PR, et al. Rubella antibody measured by radial ha-

emolysis. Characteristics and performance of a simple screening method for use in di-

agnostic laboratories. J Hyg (Camb) 1980; 84(2): 213–222.

Mace M, Cointe D, Six C, et al. Diagnostic value of reverse transcription-PCR of amniotic

fluid for prenatal diagnosis of congenital rubella infection in pregnant women with con-

firmed primary rubella infection. J Clin Microbiol 2004; 42(10): 4818–4820.


http://www.cdc.gov/nchs/data/nhanes/nhanes3/cdrom/nchs/manuals/labman.pdf

http://www.cdc.gov/nchs/data/nhanes/nhanes3/cdrom/nchs/manuals/labman.pdf

Matter L, Gorgievski-Hrisoho M, Germann D. Comparison of four enzyme immunoassays

for detection of immunoglobulin M antibodies against rubella virus. J Clin Microbiol

1994; 32: 2134–2139.

Matter L, Kogelschatz K, Germann D. Serum levels of rubella virus antibodies indicating

immunity: response to vaccination of subjects with low or undetectable antibody con-

centrations. J Infect Dis 1997; 175(4): 749–755.

McCarthy K, Taylor-Robinson CH, Pillinger SE. Isolation of rubella virus from cases in

Britain. Lancet 1963; II: 593.

McKie A, Vyse A, Maple C. Novel methods for the detection of microbial antibodies in oral

fluid. Lancet Inf Dis 2002; 2: 18–24.

Meitsch K, Enders G, Wolinsky JS, et al. The role of rubella-immunoblot and rubella-

peptide-EIA for the diagnosis of the congenital rubella syndrome during the prenatal and

newborn periods. J Med Virol 1997; 51(4): 280–283.

Meurman OH, Viljanen MK, Granfors K. Solid-phase radioimmunoassay of rubella virus

immunoglobulin M antibodies: comparison with sucrose density gradient centrifugation

test. J Clin Microbiol 1977; 5: 257–262.


successive stages of pregnancy. Lancet 1982; 2(8302): 781–784.

Miller E, Tookey P, et al. Rubella surveillance to June 1994: third joint report from the

PHLS and the National Congenital Rubella Surveillance Programme. Commun Dis Rep

1994; 4: R146–R152.

Morgan-Capner P, Crowcroft NS. PHLS Joint Working Party of the Advisory Committees

of Virology and Vaccines and Immunisation, Guidelines on the management of, and

exposure to, rash illness in pregnancy (including consideration of relevant antibody

screening programmes in pregnancy). Commun Dis Public Health 2002; 5(1): 59–71.

[http://www.hpa.org.uk/cdph/issues/CDPHvol5/No1/rash illness guidelines.pdf].

Morgan-Capner P, Miller E, Vurdien JE, Ramsay ME. Outcome of pregnancy after ma-

ternal reinfection with rubella. Commun Dis Rep 1991; 1(6): R57–R59.

Morris M, Cohen B, Andrews N, Brown D. Stability of total and rubella-specific IgG in oral

fluid samples: the effect of time and temperature. J Immunol Methods 2002; 266: 111–116.

Mortimer PP, Parry JV. Non-invasive virological diagnosis: are saliva and urine specimens

adequate substitutes for blood? Rev Med Virol 1991; 1: 73–78.

Nedeljkovic J, Jovanovic T, Mladjenovic S, et al. Immunoblot analysis of natural and

vaccine-induced IgG responses to rubella virus proteins expressed in insect cells. J Clin

Virol 1999; 14(2): 119–131.

Nokes DJ, Nigatu W, Abebe A, et al. A comparison of oral fluid and serum for the detection

of rubella-specific antibodies in a community study in Addis Ababa, Ethiopia. Trop Med

Internat Health 1998; 3: 258–267.

O’Shea S, Best JM, Banatvala JE. Viremia, virus excretion, and antibody responses after

challenge in volunteers with low levels of antibody to rubella virus. J Infect Dis 1983; 148:

639–647.

O’Shea S, Best JM, Banatvala JE. A lymphocyte transformation assay for the diagnosis of

congenital rubella. J Virol Methods 1992; 37: 139–148.

O’Shea S, Corbett KM, Barrow SM, et al. Rubella reinfection; role of neutralising antibodies

and cell-mediated immunity. Clin Diagn Virol 1994; 2: 349–358.

O’Shea S, Dunn H, Palmer S, et al. Automated rubella antibody screening: a cautionary tale.

J Med Microbiol 1999; 48: 1047.


http://www.hpa.org.uk/cdph/issues/CDPHvol5/

Parker SP, Cubitt WD. The use of the dried blood spot sample in epidemiological studies. J

Clin Pathol 1999; 52: 633–639.

Parkman PD, Buescher EL, Artenstein MS. Recovery of the rubella virus from recruits. Proc

Soc Exp Biol Med 1962; 111: 225–230.

Parkman PD, Mundon FK, McCown JM, Buescher EL. Studies of rubella. II. Neutrali-

zation of the virus. J Immunol 1964; 93: 608–617.

Perry KR, Brown WG, et al. Detection of measles, mumps and rubella antibodies in saliva

using antibody capture radioimmunoassay. J Med Virol 1993; 40: 235–240.

Pitcovski J, Shmueli E, Krispel S, Levi N. Storage of viruses on filter paper for genetic

analysis. J Virol Methods 1999; 83: 21–26.

Pustowoit B, Grangeot-Keros L, Hobman TC, Hofmann J. Evaluation of recombinant

rubella-like particles in a commercial immunoassay for the detection of anti-rubella IgG.

Clin Diagn Virol 1996; 5: 13–20.

Pustowoit B, Liebert UG. Predictive value of serological tests in rubella virus infection

during pregnancy. Intervirology 1998; 41(4, 5): 170–177.

Ramsay M, Reacher M, O’Flynn C, et al. Causes of morbilliform rash in a highly immunised

English population. Arch Dis Child 2002; 87: 202–206.

Ramsay ME, Brugha R, Brown DWG, et al. Salivary diagnosis of rubella: a study of notified

cases in the United Kingdom, 1991–4. Epidemiol Infect 1998; 120: 315–319.

Reef SE, Plotkin S, Cordero JF, et al. Preparing for elimination of congenital rubella syn-

drome (CRS): summary of a workshop on CRS elimination in the United States. Clin

Infect Dis 2000; 31: 85–95.

Revello MG, Baldanti F, Sarasini A, et al. Prenatal diagnosis of rubella virus infection by

direct detection and semiquantitation of viral RNA in clinical samples by reverse tran-

scription-PCR. J Clin Microbiol 1997; 35: 708–713.

Robert-Koch-Institut. Bekampfung der Masern und Konnatalen Roteln: WHO-Strategie in

der europaischen Region und aktueller Stand in Deutschland. Epidem Bull 2004; 10:

79–84.

Robertson SE, Featherstone DA, Gacic-Dobo M, Hersh BS. Rubella and congenital rubella

syndrome: global update. Pan Am J Public Health 2003; 14(5): 306–315.

Schmidt M, Lidqvist C, Salmi A, Oker-Blom C. Detection of rubella virus-specific immuno-

globulin M antibodies with a baculovirus-expressed E1 protein. Clin Diagn Lab Immunol

1996; 3(2): 216–218.

Skendzel LP. Rubella immunity. Defining the level of protective antibody. Am J Clin Pathol

1996; 106(2): 170–174.

Starkey WG, Newcombe J, Corbett KM, et al. Use of rubella E1 fusion proteins for the

detection of rubella antibodies. J Clin Microbiol 1995; 33: 270–274.

Stewart GL, Parkman PD, Hopps HE, et al. Rubella-virus hemagglutination-inhibition test.

N Engl J Med 1967; 276: 554–557.

Tanemura M, Suzumori K, Yagami Y, Katow S. Diagnosis of fetal rubella infection with

reverse transcription and nested polymerase chain reaction: a study of 34 cases diagnosed

in fetus. Am J Obstet Gynecol 1996; 174: 578–582.

Tang JW, Aarons E, Hesketh LM, et al. Prenatal diagnosis of congenital rubella infection in

the second trimester of pregnancy. Prenat Diagn 2003; 23: 509–512.

Tedder RS, Yao JL, Anderson MJ. The production of monoclonal antibodies to rubella

haemagglutinin and their use in antibody-capture assays for rubella-specific IgM. J Hyg

(Camb) 1982; 88(2): 335–350.


Thomas HIJ, Morgan-Capner P. Rubella-specific IgG subclass avidity ELISA and its role in

the differentiation between primary rubella and rubella reinfection. Epidemiol Infect 1988;

101: 591–598.

Thomas HIJ, Morgan-Capner P. Rubella-specific IgG1 avidity: a comparison of methods. J

Virol Methods 1991; 31: 219–228.

Thomas HIJ, Morgan-Capner P, Cradock-Watson JE, et al. Slow maturation of IgG1,

avidity and persistence of specific IgM in congenital rubella: implications for diagnosis

and immunopathology. J Med Virol 1993; 41: 196–200.

Thomas HIJ, Morgan-Capner P, Enders G, et al. Persistence of specific IgM and low avidity

specific IgG1 following primary rubella. J Virol Methods 1992a; 39: 149–155.

Thomas HIJ, Morgan-Capner P, Roberts A, Hesketh L. Persistent rubella-specific IgM

reactivity in the absence of recent primary rubella and rubella reinfection. J Med Virol

1992b; 36: 188–192.

Tipples GA, Hamkar R, Mohktari-Azad T, et al. Evaluation of rubella IgM enzyme

immunoassays. J Clin Virol 2004; 30: 233–238.

Tischer A, Gerike E. Immune response after primary and re-vaccination with different

combined vaccines against measles, mumps, rubella. Vaccine 2000; 18: 1382–1392.

Tookey PA, Cortina-Borja M, Peckham CS. Rubella susceptibility among pregnant women

in North London. J Public Health Med 2002; 24(3): 211–216.

Tookey P. Rubella in England, Scotland and Wales. Eurosurveillance Monthly 2004; 9(4):

21–22.

Tzeng WP, Zhou Y, Icenogle J, Frey TK. Novel replicon-based reporter gene assay for

detection of rubella virus in clinical specimens. J Clin Microbiol 2005; 43(2): 879–885.

Ueda K, Tokugawa K, Fukushige J, et al. Hemaglutination inhibition antibodies in con-

genital rubella syndrome. Am J Dis Child 1987; 141: 211–212.

Vaananen P, Haiva VM, Koskela P, Meurman O. Comparison of a simple latex aggluti-

nation test with hemolysis in-gel, hemagglutination inhibition, and radioimmunoassay for

detection of rubella virus antibodies. J Clin Microbiol 1985; 21(5): 793–795.

Vejtorp M, Fanoe E, Leerhoy J. Diagnosis of postnatal rubella by the enzyme-linked

immunosorbent assay for rubella IgM and IgG antibodies. Acta Pathol Microbiol 1979;

87: 155–160.

Vesikari T, Vaheri A. Rubella: a method for rapid diagnosis of a recent infection by dem-

onstration of the IgM antibodies. Br Med J 1968; 1: 221.

Vyse AJ, Brown DWG, Cohen BJ, et al. Detection of rubella virus-specific immunoglobulin

G in saliva by an amplification-based enzyme-linked immunosorbent assay using mono-

clonal antibody to fluorescein isothiocyanate. J Clin Microbiol 1999; 37: 391–395.

Vyse AJ, Cohen BJ, Ramsay ME. A comparison of oral fluid collection devices for use in the

surveillance of virus diseases in children. Public Health 2001; 115: 201–207.

Vyse AJ, Jin L. An RT-PCR assay using oral fluid samples to detect rubella virus genome for

epidemiological surveillance. Mol Cell Probes 2002; 16: 93–97.

Weller TH, Neva FA. Propagation in tissue culture of cytopathic agents form patients with


Wolinsky JS, Sukholutsky E, Moore WT, et al. An antibody- and synthetic peptide-defined

rubella virus E1 glycoprotein neutralization domain. J Virol 1993; 67(2): 961–968.

World Health Organisation. Guidelines for Surveillance of Congenital Rubella Syndrome

and Rubella. Field test version. WHO/V&B/99.22. Geneva: WHO; 1999.


World Health Organisation. Preventing congenital rubella syndrome. Wkly Epidemiol Rec

2000; 75: 290–295.

World Health Organisation. Standardization of the nomenclature for genetic characteristics

of wild-type rubella viruses. Wkly Epidemiol Rec 2005; 80: 125–132.

Zhang T, Mauracher CA, Mitchell LA, Tingle AJ. Detection of rubella virus-specific

immunoglobulin G (IgG), IgM, and IgA antibodies by immunoblot assays. J Clin

Microbiol 1992; 30(4): 824–830.

Zheng DP, Frey TK, Icenogle J, et al. Global distribution of rubella virus genotypes. Emerg

Infect Dis 2003; 9(12): 1523–1530.

Zrein M, Joncas JH, Pedneault L, et al. Comparison of a whole-virus enzyme immunoassay

(EIA) with a peptide-based EIA for detecting rubella virus immunoglobulin G antibodies

following rubella vaccination. J Clin Microbiol 1993; 31(6): 1521–1524.


Rubella Viruses



DOI 10.1016/S0168-7069(06)15004-1

79

Chapter 4

Rubella Vaccine

Susan Reef a, Stanley A. PlotkinbaRubella and Mumps Activity, National Immunization Program, Centers for DiseaseControl and Prevention, 1600 Clifton Road, Atlanta, GA 30333, USAbUniversity of Pennsylvania and Sanofi Pasteur, 4650 Wismer Road, Doylestown, PA18901, USA

Introduction

Rubella was initially thought to be one of the most benign rash illnesses. However,in 1941, Gregg associated congenital cataracts with maternal rubella, discoveringthat rubella virus was one of the most teratogenic agents known (Gregg, 1941).Twenty years later, in 1962, the rubella virus was isolated by two independentgroups, Weller and Neva at Harvard University School of Public Health andParkman, Buescher and Arenstein at Walter Reed Army Institute for Research(Parkman et al., 1962; Weller and Neva, 1962). This achievement paved the way forthe development of serological tests and vaccines. Meanwhile, a worldwide rubellapandemic, which started in Europe in 1962–1963, spread to the U.S. in 1964 and1965. The U.S. epidemic resulted in approximately 12.5 million rubella cases,11,000 fetal deaths and 20,000 congenital rubella syndrome (CRS) cases (Rubellasurveillance, 1969). This pandemic resulted in the accumulation of extensiveknowledge on rubella and CRS and reinforced the urgency of developing a rubellavaccine.

Active immunization

To develop a vaccine, researchers approached the issue in two different ways:inactivated or killed vaccine or live-attenuated virus vaccine.

dx.doi.org/10.1016/S0168-7069(06)15004-1.3d

Inactivated virus vaccine

Shortly after the isolation of the virus, investigators attempted to develop an in-activated virus vaccine, but their attempts were thwarted (Sever et al., 1963; Meyeret al., 1969). Either the vaccines were not antigenic or if antibodies were produced,it was questioned if the preparation was contaminated with live virus. Not until1980s–1990s did researchers understand the molecular and antigenic characteristicsof the rubella virus, which might serve as a basis for development of alternativevaccines (Waxman and Wolinsky, 1985). Recently, two rubella virus DNAs wereconstructed and evaluated in a mouse model. Equivalent antibody responses inmice infected with natural rubella virus and DNA vaccines were documented withpersistence of titers for 7 months (Pougatcheva et al., 1999). If developed these non-replicating rubella virus vaccines might be indicated for special circumstances, suchas vaccination of pregnant women.

Live-attenuated virus vaccine

Development

After the isolation of the rubella virus, several groups set out to develop a live-attenuated vaccine. In 1965, Parkman et al. were the first to successfully attenuatethe rubella virus with viral passage 77 times in African green monkey kidneycultures (Parkman et al., 1966). In 1969–1970, three vaccines were licensed in theU.S., including HPV-77 (dog kidney), HPV-77 (duck embryo) and Cendehill (rab-bit kidney) (Preblud et al., 1980). Shortly thereafter, the RA 27/3 vaccine (humandiploid cells) was licensed in Europe (Plotkin et al., 1969). In Japan, the initialvaccines licensed were the Takahashi (rabbit kidney) and Matsuura (Japanesequail-embryo fibroblasts) vaccines (Perkins, 1985).

Within 10 years after the initiation of the rubella vaccination program in theU.S., all three vaccines were replaced by RA 27/3, which had been licensed inEurope from 1970 onwards. In the U.S., the HPV-77 DK vaccine was withdrawndue to higher incidence of side effects as compared to the other vaccines. These sideeffects included arthritis, arthralgia and neuropathy. Shortly thereafter, the Cende-hill strain vaccine was withdrawn. In January 1979, RA 27/3 vaccine replacedHPV-77 DE vaccine and is currently the only licensed rubella vaccine in the U.S.(Preblud et al., 1980) and in most of the world.

The virus that served as the basis for RA 27/3 vaccine was isolated from the27th fetus studied by S. Plotkin during the 1964 rubella epidemic (Plotkin et al.,1969). The virus was cultured from the kidney tissue that was the third fetal ex-plant. The supernatant fluid from the kidney cells was directly planted on WI-38human diploid cells. Attenuation was accomplished by serial passage at low tem-peratures (cold adaptation). The vaccine virus was produced between the 25th and33rd passage in human diploid cells.

S. Reef, S.A. Plotkin80

After the development and licensure of the initial rubella vaccines globally,additional vaccines have been licensed in various geographic locations. In 1980, arubella vaccine (BRD-2) was developed in China using a rubella virus strain isolatedfrom a child. The rubella virus strain was isolated on human diploid cells. In a trialcomparing the BRD-2 vaccine and RA 27/3 vaccine, the seroconversion rate and mildside effects were similar (Yaru et al., 1985). In Japan, currently five different rubellavaccines are in use including TO-366 virus vaccine (Kakizawa et al., 2001).

Dose and route of administration

Worldwide, licensed rubella vaccines are administered subcutaneously with at least1000 plaque form units (PFU) at the time of delivery. During the initial clinicaltrials, subcutaneous and intranasal routes of administration were evaluated. RA27/3 vaccine was the only vaccine that produced secretory IgA, which may have arole in preventing reinfection (Hillary, 1971; Puschak et al., 1971).

Because of accelerated measles control efforts globally, various non-percuta-neous routes (e.g., intranasal, aerosol, conjunctival) of administration are beingevaluated (Cutts, 1997). A study by Dilraj documented that Edmonston–Zagreb(EZ) vaccine by aerosol produced better short term and persistence of antibodytiter when compared to EZ or Schwartz measles vaccine subcutaneously (Dilrajet al., 2000). Because of the interest of combining rubella vaccine with measlesvaccine, Mexican researchers have conducted studies aerosolizing both measles andrubella vaccine and have achieved the same seroconversion rates as after subcu-taneous injection (Sepulveda J, Valdespino JL, personal communication, 2001).However, one of the concerns about aerosolizing rubella-containing vaccine is therisk of exposing and infecting susceptible pregnant women.

Combination of the vaccine

Rubella-containing vaccines are available as a single-antigen rubella-only or in com-bination as either measles–mumps–rubella (MMR) or measles–rubella (MR) ormumps–rubella (MR). In the U.S. and increasingly elsewhere, the trivalent vaccine(MMR) is recommended for use in the pediatric schedule. Additionally, as part ofmany of the adult or childhood accelerated measles campaign, rubella vaccine hasbeen included as either MR or MMR vaccine. The MMR vaccine formulation(MMR II, Merck, Sharp and Dohme) used in the U.S. contains the Moraten-attenuated measles virus (1000 TCID50), the Jeryl Lynn mumps vaccine (5000TCID50) and the RA 27/3 rubella virus (1000 TCID50). A measles–rubella combinedvaccine (M-R-Vax II, Merck, Sharpe and Dohme) as well as measles–mumps–rubella–varicella vaccine are available.

Outside the United States, several formulations are available. In Europe andelsewhere, available formulations include Trimovax (Sanofi-Pasteur), which containsthe Schwarz measles virus (1000 TCID50), the Urabe mumps virus (20,000 TCID50)and the RA 27/3 rubella vaccine virus (1000 TCID50); and Priorix (GlaxoSmithKline),

Rubella Vaccine 81

which contains Schwartz measles virus (1000 TCID50), the RIT 4385 (derived fromJeryl Lynn, 10,000 TCID50) and the RA 27/3 rubella vaccine virus (1000 TCID50).Berna also supplies RA 27/3 rubella vaccine.

Another major manufacturer is the Serum Institute of India, which manufac-tures three different rubella-containing formulations: rubella-only vaccine (WistarRA 27/3, 1000 CCID50), MR (for export only) (L-Zagreb, 5000 CCID50) andMMR (Trestivac, EZ measles virus, 1000 CCID50).

Immune responses

Immune reactions produced by vaccination include both IgM and IgG antibodyclasses and cellular immune responses. Several different assays have been used tomeasure the immunologic response including hemagglutination assay, enzyme-linked immunoabsorbent assay (EIA) and neutralization techniques.

The hemagglutination inhibition and neutralizing responses technique wereused to measure antibody response for the earlier vaccines (HPV-77, Cendehill, RA27/3); however, as more sensitive and easier to use tests became available, they werereplaced with tests such as the EIA assay and radial diffusion. In many studies,95–100% of persons aged 11 months and older who were vaccinated with RA 27/3seroconverted (Weibel et al., 1980).

In countries that use MMR vaccine for their pediatric schedule, it is recom-mended that infants be vaccinated at 12 months or older to avoid interference bypassively acquired maternal antibodies. However, in many countries a majority ofthe mothers are now vaccinated and there is quicker loss of antibody titers ininfants of those mothers than in infants whose mothers were infected with wildrubella virus. Because of this situation, several investigators have evaluated theseroconversion rates of MMR vaccine in children o12 months of age (Singh et al.,1994; Forleo-Neto et al., 1997). These studies show that the post-vaccination se-roconversion rate is not significantly different in children aged 12 months as com-pared to children aged 15 months. However, in the Forleo-Neto study, infants aged9 months had lower post-vaccination titers to rubella than children aged 15months.

Several studies have compared antibody responses to different vaccines,particularly to those which were used earlier such as HPV-77 and Cendehill (re-viewed by Plotkin, 2004). Recent studies have been conducted comparing vaccinesincorporating rubella, mumps and measles. Priorix vaccine (GlaxoSmithKline) in-corporates the RA27/3 strain of rubella, the Schwarz strain of measles and the RIT4385 strain of mumps (related to Jeryl Lynn). Triviraten (Berna) incorporatesRA27/3, the EZ (measles) and the Rubini (mumps). When compared with MMRII, this and the Priorix vaccine had comparable antibody responses. Many antibodyassays are useful to measure titer levels; however, neutralizing antibodies may havebiological significance by measuring functionality of the antibody. In comparingdifferent vaccines, RA 27/3 vaccine induces higher neutralizing antibodies, but at alower level than natural disease (Horstmann et al., 1985).


Rubella-specific IgM antibodies can be detected within 14 days after vaccina-tion, peaking at 1 month, with variation dependent on the technique used (Enders,1985). IgM antibodies are usually gone by 3 months post-vaccination. However,using an M-antibody capture radioimmunoassay (MACRIA), some investigatorshave been able to detect low levels of rubella IgM antibodies in 18 of 53 (34%) ofvaccinees 1 year after vaccination and in 7 of 18 (39%) of vaccinees 3 years aftervaccination (Best, 1991). Rubella IgM antibodies can be detected in both primaryand reinfection; however, in reinfections, the IgM responses is transient and titersare generally lower (Balfour et al., 1981; O’Shea et al., 1994).

Viremia after vaccination occurs within 7–11 days after inoculation, but at alow level. However, pharyngeal excretion of virus is more frequent and may occurbetween 7 and 21 days. Because of the concern that the vaccine virus would betransmissible to susceptible contacts, several studies to evaluate contagiousnesswere conducted in family settings (Plotkin et al., 1969; Veronelli, 1970) and in-stitutions. There was no evidence of spread of vaccine to the susceptible contacts.In one immunization campaign with almost 100,000 children vaccinated, only oneof the 121 seronegative exposed pregnant women seroconverted (Scott and Byrne,1971). She was among the 31 pregnant women who had close contact with avaccinee. However, her elevated titer was documented only 14 days after exposureto her son who had been vaccinated, which is too soon to implicate transmission ofthe vaccine virus.

Few studies have been conducted documenting the cellular immune responsesin persons who have been vaccinated or who have had natural infection (Lallaet al., 1973; Honeyman et al., 1974; Buimovici-Klein and Cooper, 1985). Thosestudies showed that natural infection and HPV-77, Cendehill and RA 27/3 vac-cines produce cell-mediated responses with RA 27/3 being the closest to naturalinfection.

Protective effects and reinfection

The protective efficacy of rubella vaccination has been evaluated during outbreaksand by intranasal challenges of vaccinated volunteers with attenuated and unat-tenuated viruses.

The efficacy of rubella-containing vaccine has been estimated to approximately95% in outbreak settings. Additionally, when rubella vaccination was initiated dur-ing the outbreak, rubella incidence dropped 2–3 weeks later among the vaccinatedgroup (Furukawa et al., 1970; Landrigan et al., 1974). In both naturally infected andvaccinated individuals, reinfection occurs. However, the significance of reinfections ishotly debated. During an institutional outbreak of rubella, 5 of the 22 subjects (23%)previously vaccinated with HPV-77 or Benoit strains and one of the 66 (1%) nat-urally immune subjects, had subclinical reinfections without viremia (Davies et al.,1971).

Persons with natural infection are less likely to have reinfections than vaccineesand among the vaccine strains, RA 27/3 vaccine permits the fewest reinfections

Rubella Vaccine 83

(5%). In earlier studies of HPV-77 and Cendehill vaccines, the reinfection rate ofvaccinated persons was 50% or greater (Fogel et al., 1978).

Reinfection is mainly a concern when women who were either previously vac-cinated or post-natural infection have had maternal reinfection during pregnancy.In some of these situations congenital rubella was the result (Eilard andStrannegard, 1974; Forsgren et al., 1979; Bott and Eisenberg, 1982; Enders et al.,1984; Gilbert and Kudesia, 1989; Das et al., 1990; Keith, 1991; Morgan-Capner etal., 1991; Weber et al., 1993; Braun et al., 1994; Paludetto et al., 1994). However,many more women who were reinfected during pregnancy have delivered infantswithout evidence of congenital infection.

To fully understand the risk factors for reinfection, several challenge studieshave been conducted. O’Shea and colleagues measured rubella immunity in rein-fected women and seropositive volunteers who underwent a challenge with rubellavirus. The presence or absence of neutralizing antibodies or cell-mediated responsedid not correlate with reinfection (O’Shea et al., 1994). Other researchers havenoted that the risk of reinfection is correlated with lack of prior antibodies to apeptide of the E1 protein (Mitchell et al., 1996). In summary, reinfection does occurand infants with CRS have resulted from maternal reinfection, but these cases arerare compared to the number of women vaccinated who are protected.

Adverse events

The most common adverse events after vaccination include low-grade fever, rash,arthropathy and lymphadenopathy. Of all the adverse events, the occurrence ofacute arthropathy and chronic arthritis are the most concerning, particularly inpost-pubertal females.

In 1991, the Institute of Medicine reviewed data on four possible adverse eventsassociated with rubella vaccine. These included acute arthritis, chronic arthritis,neuropathies and thrombocytopenia. The committee concluded that RA 27/3 vac-cine causes acute arthritis. With regard to chronic arthritis, the committee con-cluded that, ‘‘Evidence is consistent with a causal relation between the currentlyused rubella vaccine strain (RA27/3) and chronic arthritis in adult women, al-though the evidence is limited in scope and confined to reports from one insti-tution.’’ The committee recommended that a double-blinded study be conducted toevaluate the association of chronic arthropathy associated with rubella vaccine(Howson et al., 1991).

Arthropathy

Acute arthropathy including arthritis and arthralgias may occur in up to 70% ofpersons with natural rubella, particularly post-pubertal females. Because of thefrequency of joint manifestations in natural rubella infection, initial clinical trialsevaluated and documented the frequency of acute arthropathy in those trials. Thefrequency of joint symptoms in adult females increases with age and is dependent


on which vaccine is administered. HPV-77 DK vaccine had the greatest frequencyof joint symptoms with 49% of susceptible vaccinees affected, followed by HPV-77-DE (30%), RA 27/3 (14%) and the lowest in Cendehill (9%). In either natural- andvaccine-associated disease, these symptoms usually have not caused disruptionof activities and most often have not persisted. An extensive review of acutearthropathy and arthritis associated with rubella vaccine has been previouslypublished (Howson et al., 1991).

During the mid-1980s, investigators from one institution reported persistentor chronic arthropathy in 5–11% of adult females following rubella vaccination(Tingle et al., 1986; Mitchell et al., 1993). To evaluate this finding, several re-searchers conducted studies. In a double-blinded case-control study, Slater andcolleagues compared a group of approximately 500 women who had received ru-bella vaccine post-partum with a control group of women who were immune anddid not receive rubella vaccine post-partum (Slater et al., 1995). Arthritis wasdefined as ‘‘a report of pain in any joint accompanied by at least of the following:limitation of motion, stiffness, local redness, heat or swelling: and which causedlimitation of activity or led to a professional consultation.’’ In the vaccine group, 19(3.9%) and 16 (3.2%) in the control group met the case definition for arthritis. Thedifference between the groups for arthropathy was not statistically significant.

Ray et al. conducted a retrospective cohort study comparing three differentgroups of women: seronegative immunized, seronegative unimmunized and sero-positive unimmunized (Ray et al., 1997). They found no evidence of a causalrelationship between immunization with the RA 27/3 strain of rubella virus andpersistent joint symptoms. Tingle et al. conducted a randomized double-blind, pla-cebo-controlled study enrolling 636 women (Tingle et al., 1996). Of these women, 543completed the one-month follow-up and 456 completed the 12-month follow-up. Thefrequency of acute arthropathy was 20% of the women in the placebo group and30% of the women in the vaccinated group, which was statistically significant.‘‘Persistent arthropathy was defined as occurrence of arthralgia or arthritis at anytime during the 12 month after vaccination in women who experienced acute art-hropathy and for whom joint complaints could not be attributed to other causes.’’The frequency of chronic arthropathy in the placebo group was 15% and 22% in thevaccine arm. However, whereas of 81 women in the vaccine group who developedacute arthropathy, 58 (72%) developed chronic arthropathy, 41 (75%) of the 55women in the placebo arm who developed acute arthropathy also developed chronicarthropathy. In addition, there was no description of the severity or frequency orduration of the symptoms of the women in either group. By the case definition, awoman who had an episode of acute arthralgias for 3 days and then one episode ofarthralgias of any duration after the one-month follow-up met the chronic or per-sistent arthropathy case definition. Thus, it is unclear what relationship these findingshave to the chronic arthropathy previously described. To summarize, data fromstudies in the United States and experience from other countries using the RA 27/3strain rubella vaccine have not supported an association, suggesting that chronicarthropathy may not be related to administration of rubella-containing vaccines.

Rubella Vaccine 85

Contraindications and precautions

General

Persons that have had an anaphylactic reaction following a dose of rubella-con-taining vaccine or to a vaccine component (e.g., gelatin, neomycin) should notreceive rubella-containing vaccines such as single-antigen rubella, MR or MMR(Kelso et al., 1993; Sakaguchi et al., 1995). In persons who have an egg allergy, therisk of serious allergic reactions from measles- or mumps-containing vaccines isextremely low (Greenberg and Birx, 1988; Lavi et al., 1990); however, single-antigen rubella vaccine is made in human fibroblasts cells and there is no risk forpersons with egg allergies (Plotkin, 2004).

Because of the possible risk of enhanced vaccine viral replication, persons withimmunodeficiency diseases or persons who are immunosuppressed should not bevaccinated with MMR. However, certain persons in these groups (HIV-infected,steroid treated and leukemics) may be vaccinated under certain circumstances. Inasymptomatic persons with HIV, who are severely immunocompromised and lackmeasles antibodies, MMR vaccine is recommended (CDC, 1998). Although theminimum dose or duration of treatment with steroids that causes immunosup-pression is not well defined, treatment with low dose or alternate day therapy, orwith topical or aerosol therapy, is not a contraindication to vaccination. Personswith leukemia may be vaccinated if they lack antibodies to measles, mumps orrubella and the time period between termination of therapy and vaccination is3 months or longer. Because of concern about interference with seroconversion,rubella-containing vaccine should be given either 2 weeks before, or postponed forat least 3 months after administration of blood products (e.g., immunoglobulin,whole blood).

Pregnancy

Pregnancy is a contraindication to rubella vaccine. Because of the theoretical riskto the fetus, women are counseled to avoid pregnancy for 1 month after receipt ofrubella-containing vaccine in the U.K., Canada and U.S. and 2 months in France.Because inadvertent vaccination of pregnant women does occur, many countriesestablished registries to follow the outcome of pregnancies for women vaccinated3 months prior to pregnancy or during pregnancy. In the U.S., between January1971 and April 1989, the rubella vaccination in pregnancy registry recorded 321susceptible pregnant women who delivered 324 infants. No defect associated withCRS was found. Of the 222 infants in whom serology was obtained, six (2.7%) hadsubclinical infection. All six were normal on physical examination and no defectswere identified after 2 or more years of follow-up. These data are consistent withresults reported from other countries, suggesting that if live-attenuated rubellavaccine causes defects associated with CRS, it does so at a very low rate (o1.3%)(CDC, 2001).


In 2001, the U.S. Advisory Committee on Immunization Practices recom-mended that the U.S. decrease the waiting period between vaccination and con-ception from 3 months to 1 month. This decision was based on data from the U.S.registry, unpublished data from the U.K. National Congenital Rubella SurveillanceProgramme (Tookey, 2004), Sweden and Germany (Enders et al., 1984). Data on680 infants born to susceptible women who were inadvertently vaccinated 3 monthsprior to pregnancy or during pregnancy were examined. None of the infants wasborn with CRS. Limiting the analysis to the 293 infants born to susceptible mothers1–2 weeks prior to conception and 4–6 weeks after conception, the maximumtheoretical risk was 1.3%. This risk was substantially less than the >50% risk forCRS associated with proven maternal infection during the first 20 weeks of preg-nancy (CDC, 2001). Although there has been no observed risk of CRS in infants ofmothers who were inadvertently vaccinated, Hofmann reported the first case ofpersistent fetal infection without defects following inadvertent vaccination in the 3weeks after conception. The infant had been followed more than 14 months with-out any apparent defects or late manifestations of intrauterine rubella infection(Hofmann et al., 2000).

In view of the importance of protecting women of childbearing age from rubella,reasonable practices for avoiding vaccination of pregnant women in a rubella im-munization program should include (1) asking women if they are pregnant, (2)excluding from the program those who say they are and (3) explaining the theoreticalrisks to the others before vaccinating. However, no evidence yet exists that RA 27/3damages the fetus.

Use of rubella-containing vaccine globally

Since the licensure of the rubella vaccine in 1969, the global epidemiology of rubellaand CRS has changed dramatically. Nevertheless, an estimated 100,000 cases ofCRS occur each year throughout the world (Cutts et al., 1997). In 2000, WHOconvened a meeting to review the worldwide status of CRS and its prevention(WHO, 2000). This meeting was prompted by the availability of more data on theCRS disease burden in developing countries, an increase in the number of countrieswith national rubella immunization programs and advances in laboratory diagnosissince the last international meeting on CRS and rubella in 1984. In 1996, 78 (36%)countries/territories used rubella vaccine in their national immunization programs.As of December 2003, 111 (57%) of member countries and territories used rubellavaccine. The proportion of countries using rubella vaccine varies markedly byWHO region: Africa—4%; South East Asia—18%; Eastern Mediterranean—57%;Western Pacific—52%; European—90%; and Americas—97%. The proportion ofthe world’s population living in countries with national rubella vaccination pro-grams has doubled from 12% in 1996 to 26% in 2004 (Fig. 1).

Much of the recent increase has occurred in Central and South America, inEurope and in the Western Pacific Region where polio eradication has allowednational immunization programs to address new challenges.

Rubella Vaccine 87

199665 countries12% of birth cohort

2004116 countries26% of birth cohort

Source: WHO/IVB database, 2005 and the "World Population Prospects: the 2002 Revision",New York, UN 192 WHO Member States. Data as of September 2005Date of slide: 12 September 2005

The boundaries and names shown and the designations used on this map do not imply theexpression of any opinion whatsoever on the part of the World Health Organizationconcerning the legal status of any country, territory, city or area or of its authorities, orconcerning the delimitation of its frontiers or boundaries. Dotted lines on maps representapproximate border lines for which there may not yet be full agreement.© WHO 2005. All rights reserved

Fig. 1 Countries using rubella vaccine in their national immunization system. Source: World Health Organisation, Department of Immunisation, Vaccines

and Biologicals. Reproduced with kind permission.

S.

Reef,

S.A

.P

lotk

in88

Incorporation of rubella vaccine in the national immunization program is alsorelated to a country’s economic status. In 1996, none of the 46 least developed countrieshave introduced rubella vaccine, compared with 50% (13/26) of countries with econ-omies in transition (Eastern European countries and the newly independent states ofthe former Soviet Union), 60% (53/88) of developing countries, 74% (20/27) of terri-tories, protectorates or self-governing areas and 93% (25/27) of developed countries.

Cost-effectiveness of rubella-containing vaccines in developing countries

Globally, rubella vaccination programs have not only resulted in significant re-duction of morbidity and mortality, but cost savings. Recently, Hinman et al.reviewed published information and program documents on economic analyses ofrubella and rubella containing vaccines globally (Hinman et al., 2002). Twenty-twoarticles were found, of which five cost analyses and five cost–benefit analyses (CBA)were performed in developing countries, and five CBA of rubella vaccine, five CBAof MMR vaccine and two cost-effectiveness analyses (CEA) were performed indeveloped countries.

The benefit–cost (B:C) ratio for a routine childhood programs in three developedcountries was estimated to be 5.8–11.1. One country estimated the B:C ratio to be 3.2for a two-dose program over 20 years. Of the five cost analyses in developing coun-tries, all studies were conducted in the Americas. The annual cost to treat an infantwith CRS ranged from $2,291 in Panama (1989) to $13, 482 in Jamaica, whereas inGuyana, the lifetime cost to treat an infant with CRS was $63,990.

Because of the measles elimination initiative in the Americas, many countrieshad simultaneously administered rubella-containing vaccine with the measles vac-cine as part of campaigns and routine immunization programs. CBAs attempt toplace a monetary value on the benefits of an intervention. Thus, a comparison ofthe amount spent on the intervention is divided by the cost of that intervention. Forthe entire English-speaking Caribbean to interrupt the transmission of rubella andprevent CRS, the cost–benefit ratio was 13.3:1. Two countries (Barbados, Guyana)estimated their own costs for interruption of transmission. For Barbados, the B:Cratio was 4.7:1. For Guyana, the B:C ratio was 38.8 with cost-effectiveness ratio of$1,633 per CRS case prevented.

From these studies, rubella vaccine programs globally have been cost saving.These studies are important in assisting governments to decide where to investlimited resources. When compared with other vaccine-preventable diseases such asHib and Hepatitis B, rubella vaccination programs were more cost saving. Thesestudies can assist governments to assess whether or not to implement or to enhancetheir rubella control activities.

Summary

With the introduction of rubella vaccine in 1969, significant reduction in morbidityand mortality has occurred, breaking the cycle of rubella epidemics and preventing

Rubella Vaccine 89

catastrophes like the pandemic witnessed in 1962–1965. The vaccine has proven tobe both safe and effective with persistence of effectiveness for over 30 years.

With the eradication of smallpox and near eradication of polio, the focus is nowthe elimination of measles. The significant reduction in measles by vaccination inmany countries and regions has made the burden of rubella more obvious. As aresult, over the last 10 years the use of rubella vaccine together with measles vaccinehas increased, although many populous countries are still exceptions. Nevertheless,the future looks positive for elimination from the world of rubella and its dev-astating consequences in the form of CRS.

References

Balfour Jr HH, Groth KE, Edelman CK, Amren DP, Best JM, Banatvala JE. Rubella

viraemia and antibody responses after rubella vaccination and reimmunisation. Lancet

1981; 1: 1078–1080.

Best JM. Rubella vaccines: past, present and future. Epidemiol Infect 1991; 107: 17–30.

Bott LM, Eisenberg DH. Congenital rubella after successful vaccination. Med J Aust 1982;

1: 514–515.

Braun C, Kampa D, Fressle R, et al. Congenital rubella syndrome despite repeated vac-

cination of the mother: a coincidence of vaccine failure with failure to vaccinate. Acta

Paediatr 1994; 83: 674–677.

Buimovici-Klein E, Cooper LZ. Cell-mediated immune response in rubella infections. Rev

Infect Dis 1985; 7(Suppl 1): S123–S128.

CDC. Measles, mumps and rubella—vaccine use and strategies for elimination of measles,

rubella and congenital rubella syndrome and control of mumps: recommendations of the

Advisory Committee on Immunization Practices (ACIP). MMWR 1998; 47(RR-8): 36–37.

CDC. Notice to readers: revised ACIP recommendations for avoiding pregnancy after re-

ceiving a rubella-containing vaccine. MMWR 2001; 50: 1117.

Cutts FT, Robertson SE, Diaz-Ortega JL, Samuel R. Control of rubella and congenital

rubella syndrome (CRS) in developing countries, Part 1: Burden of disease from CRS.

Bull World Health Organ 1997; 75: 55–68.

Das BD, Lakhani P, Kurtz JB, et al. Congenital rubella after previous maternal immunity.

Arch Dis Child 1990; 65: 545–546.

Davies WJ, Larson HE, Simsarian JP, et al. A study of rubella immunity and resistance to

infection. JAMA 1971; 215: 600–608.

Dilraj A, Cutts FT, Bennett JV, Fernandez de Castro J, Cohen B, Coovadia HM. Persistence

of measles antibody two years after revaccination by aerosol or subcutaneous routes.

Pediatr Infect Dis J 2000; 19(12): 1211–1213.

Eilard T, Strannegard O. Rubella reinfection in pregnancy followed by transmission to the

fetus. J Infect Dis 1974; 129: 594–596.

Enders G, Calm A, Schaub J. Rubella embryopathy after previous maternal rubella

vaccination. Infection 1984; 12: 96–98.

Enders G. Serologic test combinations for safe detection of rubella infections.

Rev Infect Dis 1985; 7(Suppl 1): S113–S122.


Fogel A, Gerichter CB, Barena B, et al. Response to experimental challenge in persons

immunized with different rubella vaccines. J Pediatr 1978; 92: 26–29.

Forleo-Neto E, Carvalho ES, Fuentes IC, Precivale MS, Forleo LH, Farhat CK. Serocon-

version of a trivalent measles, mumps, and rubella vaccine in children aged 9 and 15

months. Vaccine 1997; 15: 1898–1901.

Forsgren M, Calstrom G, Strangert K. Congenital rubella after maternal reinfection. Scand

J Infect Dis 1979; 11: 81–83.

Furukawa T, Miyata T, Konda K, et al. Rubella vaccination during an epidemic. JAMA

1970; 213: 987–990.

Gilbert J, Kudesia G. Fetal infection after maternal reinfection with rubella. BMJ 1989; 299:

12–17.

Greenberg MA, Birx DL. Safe administration of mumps–measles–rubella vaccine in egg-

allergic children. J Pediatr 1988; 13: 504–506.

Gregg NM. Congenital cataract following German measles in the mother. Trans Ophthalmol

Soc Aust 1941; 3: 35–46.

Hillary IB. Trials of intranasally administered rubella vaccine. J Hyg (Camb) 1971; 69:

547–552.

Hinman AR, Irons B, Lewis M, Kandola K. Economic analyses of rubella and rubella

vaccines: a global review. Bull World Health Organ 2002; 80: 264–270.

Hofmann J, Kortung M, Pustowoit B, et al. Persistent fetal rubella vaccine virus infec-

tion following inadvertent vaccination during early pregnancy. J Med Virol 2000; 61:

155–158.

HoneymanMC, Forrest JA, Dorman DC. Cell-mediated immune response following natural

rubella and vaccination. Clin Exp Immunol 1974; 17: 665–671.

Horstmann DM, Schluederberg A, Emmons JE, Evans BK, Randolph MF, Andiman WA.

Persistence of vaccine-induced immune responses to rubella: comparison with natural

infection. Rev Infect Dis 1985; 7(Suppl 1): S80–S85.

Howson CP, Howe CJ, Fineberg HV. Chronic Arthritis in Adverse Effects of Pertussis and

Rubella Vaccines. Washington, DC: Institute of Medicine, National Academy; 1991; pp.

187–205.

Kakizawa J, Nitta Y, Yamashita T, Ushijima H, Katow S. Mutations of rubella virus

vaccine TO-336 strain occurred in the attenuation process of wild progenitor virus. Vac-

cine 2001; 19: 2793–2802.

Keith CG. Congenital rubella infection from reinfection of previously immunized mothers.

Aust N Z J Ophthalmol 1991; 19: 291–293.

Kelso JM, Jones RT, Yunginger JW. Anaphylaxis to measles, mumps, and rubella vaccine

medicated by IgE to gelatin. J Allergy Infect Dis 1993; 91: 867–872.

Lalla M, Vesikari T, Virolainen M. Lymphoblast proliferation and humoral antibody

response after rubella vaccination. Clin Exp Immunol 1973; 15: 192–202.

Landrigan PL, Stoffels MA, Anderson E, Witte JJ. Epidemic rubella in adolescent boys:

clinical features and results of vaccination. JAMA 1974; 227: 1283–1288.

Lavi S, Zimmermann B, Koren G, Gold R. Administration of measles, mumps, and rubella

virus vaccine (live) to egg-allergic children. JAMA 1990; 263: 269–271.

Meyer Jr HM, Parkman PD, Hobbins TE, Larsons E, Davis WJ, Simsarian JP, Hopps HE.

Attenuated rubella virus: laboratory and clinical characteristics. Am J Dis Child 1969;

118: 155–163.

Rubella Vaccine 91

Mitchell L, Ho M, Rogers J, et al. Rubella reimmunization: comparative analysis of the

immunoglobulin G response to rubella virus vaccine in previously seronegative and

seropositive individuals. J Clin Microbiol 1996; 34: 2210–2218.

Mitchell LA, Tingle AJ, Shukin R, Sangeorzan JA, McCune J, Braun DK. Chronic rubella

vaccine-associated arthropathy. Arch Intern Med 1993; 153: 2268–2274.

Morgan-Capner P, Miller E, Vurdien J, Ramsay M. Outcome of pregnancy after maternal

reinfection with rubella. Commun Dis Rep 1991; 1: R57–R59.

O’Shea S, Corbett K, Barrow S, et al. Rubella reinfection: role of neutralizing antibodies and

cell-mediated immunity. Clin Diagn Virol 1994; 2: 349–358.

Paludetto R, van den Huevel J, Stagni A, et al. Rubella embrypopathy after maternal

infection. Biol Neonate 1994; 65: 340–341.



Parkman PD, Meyer HM, Kirschstein BL, Hopps HE. Attenuated rubella virus. I. Devel-

opment and laboratory characterization. N Engl J Med 1966; 275: 569–574.

Perkins FT. Licensed vaccines. Rev Infect Dis 1985; 7(Suppl 1): S73–S76.

Plotkin SA, Farquahr JD, Katz M, Buser F. Attenuation of RA 27/3 rubella virus in WI-38

human diploid cells. Am J Dis Child 1969; 118: 178–185.

Plotkin SA. Rubella vaccine. In: Vaccines (Plotkin SA, Orenstein WA, editors). 4th ed.,

Chapter 26. Philadelphia: W.B. Saunders Co.; 2004; pp. 703–743.

Pougatcheva SO, Abernathy ES, Vzorov AN, Compans RW, Frey TK. Development of a

rubella virus DNA vaccine. Vaccine 1999; 17: 2104–2112.

Preblud SR, Serdula MK, Frank Jr JA, Brandling-Bennett AD, Hinman AR. Rubella vac-

cination in the United States: a ten-year review. Epidemiol Rev 1980; 2: 171–194.

Puschak R, Young M, McKee TV, Plotkin SA. Intranasal vaccination with RA 27/3

attenuated rubella virus. J Pediatr 1971; 79: 55–70.

Ray P, Black S, Shinefield H, et al. Risk of chronic arthropathy among women after rubella

vaccination. JAMA 1997; 278: 551–556.

Rubella Surveillance. National Communicable Disease Center, United States Department of

Health, Education and Welfare (No.1); June 1969.

Sakaguchi M, Ogura H, Inouye S. IgE antibody to gelatin in children with immediate-type

reactions to measles and mumps vaccines. J Allergy Infect Dis 1995; 96: 563–565.

Scott HD, Byrne EB. Exposure of susceptible pregnant women to rubella vaccines: sero-

logical findings during the Rhode Island immunization campaign. JAMA 1971; 215:

609–612.

Sepulveda J, Valdespino JL, personal communication 2001.

Sever JL, Schiff GM, Huebner RJ. Inactivated rubella virus vaccine. J Lab Clin Med 1963;

62: 1015.

Singh R, John TJ, Cherian T, Raghupathy P. Immune response to measles, mumps & rubella

vaccine at 9, 12 & 15 months of age. Indian J Med Res 1994; 100: 155–159.

Slater P, Ben-Zvi T, Fogel A, et al. Absence of an association between rubella vaccination

and arthritis in underimmune post-partum women. Vaccine 1995; 13: 1529–1532.

Tingle A, Mitchell L, Grace M, et al. Randomised double-blind placebo-controlled study on

adverse effects of rubella immunization in sero-negative women. Lancet 1996; 349:

1277–1281.


Tingle AJ, Allen M, Petty RE, Kettyls GD, Chantler JK. Rubella-associated arthritis.

I. Comparative study of joint manifestations associated with natural rubella infection and

RA 27/3 rubella immunization. Ann Rheum Dis 1986; 45: 110–114.

Tookey P. Rubella in England, Scotland and Wales. Euro Surveill 2004; 9: 21–23.

Veronelli JA. An open community trial of live rubella vaccines. Study of vaccine virus

transmissibility and antigenic efficacy of three HPV-77 derivatives. JAMA 1970; 213:

1829–1836.

Waxman MN, Wolinsky JS. A model of the structural organization of rubella virions. Rev

Infect Dis 1985; 7(Suppl 1): 133–139.

Weber B, Enders G, Schlosser R, et al. Congenital rubella syndrome after maternal rein-

fection. Infection 1993; 21: 118–121.

Weibel RE, Villarejos VM, Klein EB. Clinical laboratory studies of live attenuated RA 27/3

and HPV-77 DE rubella virus vaccine. Proc Soc Exp Biol Med 1980; 165: 44–49.

Weller TH, Neva FA. Propagation in tissue culture of cytopathic agents from patients with


WHO. Report of a Meeting on Preventing Congenital Rubella Syndrome: Immunization

Strategies, Surveillance Needs. Geneva: WHO/V&B/00.10; 12–14 January 2000.

Yaru H, Zhao K, Yinxiang G, Sulan H, Shuzhen W, Changtai W. Rubella vaccine in the

People’s Republic of China. Rev Infect Dis 1985; 7(Suppl 1): S79.

Rubella Vaccine 93

Rubella Viruses



DOI 10.1016/S0168-7069(06)15005-3

95

Chapter 5

Rubella Epidemiology: Surveillance toMonitor and Evaluate Congenital RubellaPrevention Strategies

Catherine Peckham, Pat Tookey, Pia HardelidCentre for Paediatric Epidemiology and Biostatistics, UCL Institute of Child Health,30 Guilford Street, London, WC1N 1EH, UK

Rubella infection usually presents as a mild or asymptomatic infection in childrenand adults and often goes undiagnosed. In pregnant women however, the conse-quences can be serious, especially when infection occurs in the first trimester. Thevirus is teratogenic and may cross the placenta causing fetal infection, which ifacquired early in fetal life may be associated with serious consequences. The mostcommonly described congenital rubella anomalies include sensorineural hearingimpairment, cataracts, cardiac defects, impaired fetal growth and mental retarda-tion (see Chapter 2). The aim of vaccination programmes is therefore to preventcases of congenital rubella. To effectively achieve this goal, comprehensivesurveillance programmes to monitor immunity in the population and detect casesof both rubella and congenital rubella are needed along with an understanding ofthe likely outcomes of different vaccination strategies, which is not intuitive.

Congenital rubella infection

There is a paucity of reliable information on the prevalence of congenital rubelladefects and disability, particularly in developing countries, due to the absence ofroutine congenital rubella surveillance and the difficulty of diagnosing congenitalrubella as the cause of a defect after the first months of life. However, at least100,000 or more cases occur each year worldwide (World Health Organisation,Department of Vaccines and Biologicals, 2000). Only a minority of affected infantspresent with the classic triad of clinical signs that are recognisable at birth, which is

dx.doi.org/10.1016/S0168-7069(06)15005-3.3d

often referred to as congenital rubella syndrome (CRS). A retrospective diagnosisof congenital rubella is not possible when a child presents later in childhood with adefect compatible with congenital rubella, such as sensorineural hearing loss, ascongenital infection cannot be distinguished from natural infection or vaccine-induced immunity. This is particularly so in a non-epidemic situation. Nevertheless,studies have shown that the incidence of congenital rubella defects in developingcountries is at least as high as that reported in developed countries before theintroduction of vaccine. Most estimates of the prevalence of congenital rubelladefects following congenital infection are based on the outcome of pregnanciesidentified during a rubella epidemic. These studies may underestimate the risk ofadverse effects as follow up is usually short term and isolated cases of sensorineuralhearing loss are not likely to be identified. Based on a comprehensive review of theliterature, the incidence of congenital rubella in developing countries during epi-demics has been estimated to be 0.5–2.2 per 1000 live births (about 1.7 per 1000 livebirths in Israel, 0.7 in Oman, 2.2 in Panama, 1.5 in Singapore and 1.7 in Jamaica)(Cutts et al., 1997). These rates are comparable to those in industrialised countriesin the pre-vaccine era. It is logical to assume therefore that the problem of con-genital rubella is universal and that where rubella susceptibility in women of child-bearing age is high, rubella poses a serious problem.

In most countries data on the prevalence of congenital rubella defects in non-epidemic situations remains limited. Epidemiological studies carried out in the UKin the pre-vaccine era suggested that congenital rubella accounted for at least 16%of the cases of moderate to severe sensorineural deafness (Peckham et al., 1979) andabout 2% of the cases of congenital heart disease among British children(Peckham, 1985). In France, it was estimated that 16% of congenital cataracts inchildren could be attributed to congenital rubella (Celers et al., 1983). Studies indeveloping countries, which have largely focused on specific conditions such ascongenital cataracts or childhood deafness, have identified congenital rubella as animportant potentially preventable cause of these defects (Eckstein et al., 1996).These studies highlight the need for establishing rubella vaccination programmesthroughout the world.

Risk of congenital infection and risk of defects

When rubella infection is acquired in early pregnancy the risk of fetal infection withassociated defects is high and a substantial number of infants are severely affectedwith multiple defects, particularly of the eye, heart and ear and neurologicalabnormalities (see Chapter 2). In contrast, fetal infection later in the third and fourthmonth usually results in sensorineural deafness alone. Most congenitally infectedinfants are affected as a result of primary maternal infection although damageresulting from asymptomatic or clinically apparent maternal reinfection in women

C. Peckham et al.96

with previous immunity has also been described (Miller, 1990; Morgan-Capner etal., 1991; Ushida et al., 2003). The risk of fetal infection following maternal rein-fection during the first 12 weeks of pregnancy is thought to be low, probably lessthan 10% (Miller, 1990; Morgan-Capner et al., 1991); fetal infection is not likely tobe associated with damage following subclinical maternal reinfection, but the riskmight be higher following symptomatic reinfection, which is very rare.

In a large prospective study in the UK the pregnancy outcome of over 1000women with laboratory confirmed rubella infection in pregnancy was established.However, over 90% of women infected in the first trimester of pregnancy and halfof those with infection between 13 and 16 weeks of gestation chose to have atherapeutic abortion. The dates of the last menstrual period and rash were known,and the infants were followed up systematically for several years (Miller, 1991).Studies based on serological diagnosis of rubella in pregnancy with longer termfollow-up are more likely to reflect the true risk of damage than those earlier studiesbased on a clinical diagnosis. Table 1 shows the risk of fetal infection by week ofmaternal infection, the risk of fetal damage in children with proven infection andthe overall risk of damage in a pregnancy complicated by rubella.

Overall, around 80% of infants exposed to maternal rubella in the first 8 weeksof pregnancy had severe rubella-associated impairment including congenital heartdefects, eye defects and hearing impairment.

Congenital rubella defects occurred in about 50% of infants exposed tomaternal rubella in weeks 9–12 and at this stage hearing impairment was the mostfrequent defect. Exposure to maternal rubella at 13–16 weeks resulted in fewer than30% of infants with damage and defects other than congenital hearing loss wereunusual. Infections after 16 weeks of gestation were only occasionally associatedwith defects, although cases of deafness have been reported up to 22 weeks ofgestation. Well-designed prospective studies of children born after later infectionhave shown that although rubella infection after this period can result in fetalinfection, there appears to be no increased risk of defects (Miller et al., 1982;Grillner et al., 1983).

Table 1

Risk of infection, defect and overall risk of defects following serologically confirmed rubella at successive

stages in pregnancy

Week Risk of infection (%) Risk of defect (%) Overall risk (%)

2–10 100 89 89

11–12 73 50 37

13–16 52 33 17

17–18 38 7 3

19+ 25–82 0 0

Source: Data from Miller et al. (1982, 1991).

Rubella Epidemiology 97

Preconception rubella

Although there have been isolated reports of possible congenital defects followingmaternal rubella before conception, in none of these cases was the maternal or fetalinfection confirmed by adequate laboratory tests. In a prospective study, Enderset al. (1988) found no risk associated with exposure to rubella before conception.

Long-term manifestations of congenital rubella

It is difficult to establish the full impact of congenital rubella on a population, asthere is not always a history of maternal infection and defects may not be apparentor even develop until weeks, months or years after birth, when a retrospectivediagnosis of congenital infection is usually no longer possible (Peckham, 1972).Sensorineural hearing loss, the most frequent late-onset defect, may be moderate orsevere, bilateral or unilateral and is often associated with pigmentary retinopathy.Other manifestations of late-onset disease include pneumonitis, diabetes mellitus,hypothyroidism, growth hormone deficiency and encephalitis (Marshall, 1972).Follow up of established cohorts of children with congenital rubella diagnosed atbirth into adult life have demonstrated the progressive nature of the disease and itslong-term sequelae. The follow-up in 1990 of children with congenital rubella bornin the 1970s and 1980s and reported to the UK National Congenital RubellaSurveillance Programme demonstrated an increased incidence of diabetes and thy-roid problems in adolescents with congenital rubella compared with the generalpopulation. This concurs with findings reported from the original Australian co-hort where follow-up to age 60 years has been reported (Menser et al., 1967, 1978;Forrest et al., 1971; McIntosh and Menser, 1992; Forrest et al., 2002). There arereports of higher than expected rates of hearing loss, glaucoma, insulin-dependentdiabetes, hypo- and hyperthyroidism, hormone imbalances, premature ageing andgastrointestinal and oesophageal problems, but these were mostly based on follow-up of small numbers of individuals with problems already manifested in early life.There is a paucity of information on the later development of children with con-genital infection and no signs of rubella damage in early life. It has been suggestedthat an autoimmune process may play a role in some of these late-onset conditions(Cooper and Alford, 2001).

Epidemiological features of the rubella virus

The widespread use of rubella vaccine has had a marked impact on the rubellaantibody profile of the population and the prevention of congenital rubella. In mostcountries, in the absence of rubella immunisation, rubella is endemic with epidemiccycles occurring every 4–7 years, usually in the spring or early summer in temperateclimates, and the size of an outbreak depends on the number of susceptible in-dividuals in the population and their clustering within the population. Seroepide-miological studies carried out in industrialised countries in the pre-vaccination era

C. Peckham et al.98

generally showed that the proportion of individuals with rubella antibodiesincreased steadily with age and that overall, about 80% of women of childbearingage had rubella antibodies and around 20% were still susceptible to infection. Mostinfections were acquired early in childhood, with around 50% of childrenseropositive by 6–8 years and a peak age of infection between 4 and 11 years.However, in some countries the rate of acquisition of antibodies was more rapidthan in others, probably due to factors relating to crowding, social mixing and todifferent patterns of childcare. There were some notable exceptions where suscep-tibility rates remained high, including parts of Asia and Africa and some islandpopulations (Miller, 1991). In these situations, the introduction of wild rubellavirus into the community would result in rubella epidemics with serious conse-quences for pregnant women and their offspring.

Rubella susceptibility among pregnant women

The proportion of susceptible women, and hence the risk of rubella in pregnancy,may vary widely between population subsets within a country as well as betweencountries. For example, in a population-based seroprevalence study in the Neth-erlands blood samples were collected from over 8000 individuals, 20% of whomlived in municipalities with low uptake of vaccine. Serological profiles of these data,which were representative of the population, showed clusters of susceptible indi-viduals within specific age, social and geographical groups (de Haas et al., 1999). Amore recent study in the Netherlands demonstrated that although overall vacci-nation rates were high and herd immunity sufficient among the general population,this was not the case in orthodox reformed communities (de Melker et al., 2003). In2004, a rubella outbreak within these communities led to several children beingborn with congenital rubella-related defects (van der Veen et al., 2005). In the UK,in the 1980s, a disproportionate number of congenitally infected infants were bornto Asian women, a group known to have higher rates of rubella susceptibility thannon-Asian women (Miller et al., 1991). In contrast, women most affected in the1960s were recent immigrants from the Caribbean, where there had not been arubella epidemic for many years and the prevalence of seropositivity was low.

More than 25 years after the introduction of rubella immunisation, analysis ofblood samples collected routinely from women receiving antenatal care in NorthLondon showed that women from minority ethnic groups in the UK were morelikely to be susceptible to rubella than white women (Table 2), a finding likely toreflect recent patterns of immigration and missed rubella vaccination in childhood;primiparous women were particularly likely to be susceptible (Tookey et al., 2002).

In these circumstances, if exposed to circulating rubella, women from minorityethnic groups are at increased risk of contracting rubella in pregnancy and having achild with congenital rubella. In the US, rubella now occurs mainly among foreignborn Hispanic adults who are either unvaccinated or whose vaccination status isnot known, and there is very little spread among the US resident population (Reefet al., 2002). These observations highlight the need for monitoring rubella


seroprevalence within a country so that groups at risk of rubella infection in preg-nancy, such as immigrants and refugees from countries with no immunisationprogrammes, can be identified and targeted immunisation programmes introduced.

Studies among pregnant women in developing countries have shown widevariation in the age-specific seroprevalence of rubella ranging from 2.7% to 68%.In an analysis of serological data from 45 developing countries the proportion ofseronegative women was >25% in 12 countries, 10–24% in 20 countries ando10% in 13 countries (Cutts and Vynnycky, 1999).

Prevention of congenital rubella

Rubella vaccines

The last major epidemic in the US occurred in 1963–1964 and resulted in anestimated 13,000 fetal or neonatal deaths and 20,000 children born with congenitalrubella-related defects (Cooper and Alford, 2001). This epidemic had enormoushuman and economic cost implications and was a major impetus for the deve-lopment of a rubella vaccine and for the rapid implementation of vaccinationprogrammes.

Table 2

Proportion of pregnant women susceptible to rubella by ethnic group and parity, North London

1996–1999

Ethnic group Total number tested Susceptible

women (%)

Susceptible primiparous

women (%)

White British 94,626 1.6 1.9

Other White 2480 3.4 3.8

Mediterranean 3715 3.7 4.4

Pakistani 4312 4.1 5.4

Indian 8469 4.4 6.7

Bangladeshi 2171 6.1 9.4

Sri Lankan 763 14.9 23.3

Other Asian 955 5.9 10.4

Oriental 1896 8.0 9.4

Black British 240 2.1 2.0

Black Caribbean 3242 3.1 3.2

Black African 4878 6.2 8.4

Other Black 651 4.3 6.5

Source: Data from Tookey et al. (2002).

C. Peckham et al.100

The primary purpose of rubella vaccination is to prevent congenital rubellainfection (see Chapter 4). The first live-attenuated rubella vaccines were licensed foruse in 1969 and with the introduction of rubella vaccine into national immunisationprogrammes congenital rubella became a preventable condition. Rubella vaccine issafe and highly effective with a vaccine efficacy of about 98%. The vaccine can bedelivered in combination with measles and mumps vaccine as MMR vaccine, in theroutine childhood immunisation programme.

Rubella vaccine in pregnancy

Rubella vaccine given during or shortly before conception carries a theoretical riskof congenital rubella infection. Rubella vaccine virus has been shown to cross theplacenta and to infect the fetus, and vaccine virus has been isolated from productsof conception. Rubella-specific IgM, which indicates congenital infection, is presentin about 5% of infants born to rubella-susceptible mothers vaccinated around thetime of conception. In the UK, 4 of 25 infants born to susceptible women, whowere vaccinated more than 1 week after the last menstrual period and followedprospectively, had rubella IgM at birth, but none of 21 infants born followingvaccination of their susceptible mothers up to 3 months before LMP. None of theseinfants, including the four with rubella IgM, had evidence of congenital rubelladamage (Tookey, 2001). In a similar study in Germany, in 1 of 5 cases, vaccinevirus was transmitted to the infant as evidenced by persistent fetal infection. In onecase, virus shedding persisted for more than 8 months and sequence analysis carriedout on virus isolates from amniotic fluid, cord blood and the infant’s urine con-firmed infection by the vaccine strain; the infant had no evidence of rubella defects(Hofmann et al., 2000). These results have been collated with findings fromprospective studies in various countries and no infants born to seronegative womenwho received rubella vaccine within 3 months of conception have been identifiedwith congenital rubella defects (Best and Banatvala, 2004). Based on data fromthese studies and using the upper 95% confidence limit the maximum theoreticalrisk of a woman receiving rubella vaccine up to 3 months before conception givingbirth to a child with congenital rubella defects is o0.6% for women vaccinatedwith Cendehill or RA 27/3 rubella vaccine strains. When the focus was limited towomen vaccinated with RA 27/3 rubella vaccine strain vaccine between 1 and 6weeks after the last menstrual period, the maximum theoretical risk for susceptiblewomen is o2.5%, but numbers are small and confidence limits wide. Follow up ofinfants exposed to rubella vaccination close to the time of conception has usuallybeen short and has focused on birth outcome when only the more serious defectsare likely to have been identified (Best and Banatvala, 2004). Nevertheless, basedon a review of all available information, inadvertent rubella vaccination inpregnancy is no longer considered to be an indication for termination of pregnancy.


Vaccination strategies—theoretical and practical considerations

Several research groups have modelled the transmission dynamics of rubella virusto assess the impact of different vaccine strategies and levels of vaccination uptakeon age at infection, population seroprevalence and the number of pregnanciesaffected by rubella. Many studies focus on the dichotomy between schoolgirl vacci-nation programmes, sometimes referred to as a selective vaccination strategy, andthe vaccination of boys and girls in early life, i.e. a universal strategy (Knox, 1985).A selective vaccination strategy can never completely prevent all cases of congenitalrubella infection, since virus transmission is maintained among young children.However, introducing a selective strategy will always serve to decrease the numberof cases of congenital rubella infection. This is in contrast to the universal strategy,which interrupts transmission of infection among children. In the long term, highrates of vaccination coverage will result in a population with few susceptible in-dividuals and an interruption of rubella virus circulation, thus protecting pregnantwomen from infection. However, if high vaccine levels are not maintained, then thiscan result in a reduced circulation of the virus thereby delaying the age at whichinfection occurs in susceptible (unvaccinated) individuals, which may lead to anincrease in the number of cases of congenital rubella (Anderson and May, 1983;Knox, 1985; Anderson and Grenfell, 1986; Anderson and May, 1991). This isparticularly the case in countries with an underlying low rate of trans-mission, which leaves a significant proportion of women susceptible in childbear-ing age. This shift in the age at risk was clearly demonstrated in recent rubellaoutbreaks in Brazil and Costa Rica (Castillo-Solorzano et al., 2003).

Assuming a vaccine effectiveness of over 95% and homogenous mixing, it isestimated that a vaccination coverage of at least 85% is required to achieve acritical level of immunity to prevent the spread of infection within the population(Anderson and May, 1983; Nokes and Anderson, 1988). However, even when highoverall vaccine coverage is achieved, herd immunity can be compromised if thereare groups of unvaccinated individuals within a community, clustered geograph-ically on religious or socio-demographical grounds. Outbreaks of measles, mumpsand rubella and clusters of congenital rubella cases have been documented in thesesituations (Briss et al., 1992; Mellinger et al., 1995; Hanratty et al., 2000; van derVeen et al., 2005).

This possible adverse effect of the introduction of a universal rubella vacci-nation programme needs to be taken into account when planning congenital ru-bella prevention strategies. In countries such as Ethiopia, where infection rates arehigh among young children, only a small proportion of women of childbearing ageremain susceptible (Cutts et al., 2000). The introduction of rubella vaccine wouldnot be appropriate unless very high levels of vaccine coverage could be maintained,which is unlikely in this and many other resource-poor settings, as it could result inan increase in the incidence of congenital rubella defects.

Despite recommendations made by the WHO to set up integrated nationalsurveillance systems to monitor cases of rubella and congenital rubella, as well as


the proportion of women of reproductive age who are seronegative (Cutts et al.,1999), few countries have an active congenital rubella surveillance system in placeand given the inadequate uptake of MMR vaccination in many European countriesthe eradication of congenital rubella is unlikely to be achieved in the near future. Insome countries, such as Italy and Germany, MMR levels of vaccine uptake are notsufficient to interrupt rubella transmission in the community. In the UK, the recentdecline in uptake of both doses of MMR vaccine (illustrated with data for Englandin Table 3) is also a cause for concern, particularly in London, where immunisationuptake for the 1st dose of MMR is only slightly higher than 50% in some localities(NHS Health and Social Care Information Centre, 2005).

Outbreaks of measles have already been reported in the UK (Jansen et al., 2003;Gupta et al., 2005) and across Europe. In Ireland, a measles outbreak in Dublin in2000 led to 1115 reported cases, 355 hospital admissions and 3 deaths (McBrienet al., 2003). In early 2006, several measles outbreaks were reported in England,which led to hospitalisations and caused one death (Health Protection Agency,2006). Several recent rubella epidemics have also been reported across Europe. Inaddition to the outbreak in the Netherlands mentioned above, an outbreak inRomania in 2002–2003 led to 115 000 reported cases (approximately 531/100 000population) as well as 150 children born with suspected rubella-associated defects.Seven of these were laboratory confirmed (Rafila et al., 2004). These examplesdemonstrate that rubella virus is still circulating in Europe and can lead to out-breaks among unvaccinated groups. Although a single dose of vaccine with highcoverage will produce prolonged periods of freedom from rubella, cohorts ofsusceptible individuals who did not receive vaccine or who failed to respond to a

Table 3

MMR uptake in England, by year of birthday

Year 1st dose by 2nd

birthday (%)

1st dose by 5th

birthday (%)

1st and 2nd dose by

5th birthday (%)

1994–1995 91

1995–1996 92

1996–1997 92

1997–1998 91

1998–1999 88

1999–2000 88 93 76

2000–2001 87 92 75

2001–2002 84 91 74

2002–2003 82 90 75

2003–2004 80 90 75

2004–2005 81 89 73

Source: NHS Immunisation Statistics, England: 2004–2005 (NHS Health and Social Care Information

Centre, 2005).


single dose, will accumulate over time, along with the risk of future epidemics,usually among older children and young adults. In view of this, all countries in theEuropean region of the WHO now recommend two doses of MMR (WeeklyEpidemiological Record, 2005).

There are few studies documenting waning immunity following vaccination orinfection. However, with the diminished circulation of virus within the communityand the lack of opportunity for immunological boosting of either vaccine or naturalimmunity through repeated exposure to infection, natural immunity may not belifelong and vaccine-induced immunity may be lost even more rapidly than thatresulting from natural infection (Davidkin et al., 2000). More women could then besusceptible to symptomatic reinfection and lower levels of maternal immunitycould result in shorter duration of passive immunity in the infant. Seroprevalencestudies are required to monitor age-specific rubella antibodies so that any decline inpopulation immunity can be detected as early as possible. In a Swedish study,where a cohort of children aged 12 years was followed up for 8–16 years to monitorrubella immunity and seroconversions after the introduction of MMR, there wasevidence of a loss of immunity with natural immunity persisting at higher levelsthan vaccine-induced immunity (Christenson and Bottiger, 1994).

The implementation of rubella vaccination programmes: global experiences

The first step in eliminating rubella infection is to build up a picture of the ep-idemiology of the infection through clinical and laboratory surveillance, and todetermine the feasibility of achieving and maintaining high vaccine coverage tointerrupt transmission while avoiding the risk of shifting the age of acquisition ofrubella infection to older age groups. Targets for vaccination coverage need to beset. There also needs to be an assessment of the age- and sex-specific seroprevalenceof rubella antibodies in the population, and where possible an estimate of theincidence of congenital rubella abnormalities. Inadequate use of rubella vaccine cando more harm than good; countries attempting rubella elimination should ensurethat routine vaccine coverage in children is higher than 85% and sustainable in thelong term, and that women of childbearing age are immune. Immunisation ofchildren will alter the transmission dynamics and could lead to an increase insusceptibility in young women with the potential for an increase in the incidence ofcongenital rubella (Vynnycky et al., 2003).

Most rubella vaccination programmes were introduced following epidemics ofrubella, which had highlighted the severity of the problem and heightened publicand political awareness. In Poland, a rubella epidemic was reported in 1985–1986 inwhich many pregnant women were exposed and in 1988 rubella vaccine wasintroduced for 13-year-old schoolgirls (Zgorniak-Nowosielska et al., 1996). Theinitial approach, adopted by many countries, has been to immunise specific groupsof individuals at risk of acquiring rubella in pregnancy, such as adolescent girls andrubella-susceptible women of childbearing age. The aim of this approach was to‘mop up’ those who had escaped childhood infection rather than to interrupt


transmission of infection within the population. It also allowed for the boosting ofnatural and vaccine-induced antibody by repeated exposure to infection therebyavoiding the potential risk of waning immunity in a vaccinated population. Thisapproach had an immediate effect in reducing congenital rubella infection and incountries such as the UK, Australia and Canada was followed by the later intro-duction of universal immunisation for all children with MMR vaccine.

In the UK, rubella vaccination was introduced in 1970 for schoolgirls between11 and 14 years of age. This was soon extended to include susceptible women ofchildbearing age, particularly those working with children. Pregnant women foundto be susceptible to rubella at antenatal screening were offered vaccine postpartum.These policies resulted in a reduction in notifications of congenital rubella andrubella-associated terminations of pregnancy. Following the introduction in 1988of MMR vaccine for both boys and girls in their second year of life, the 1994 masscampaign to immunise all school children aged 5–16 years with measles/rubellavaccine, and the addition of a second dose of MMR for pre-school children from1996, circulation of wild rubella virus declined, and congenital rubella was virtuallyeliminated (Miller et al., 1997; Tookey and Peckham, 1999; Tookey 2004). Theschoolgirl vaccination programme was discontinued in 1996, but routine antenataltesting for rubella antibody status with vaccination of susceptible women in thepostpartum period is still recommended (HMSO, 1996).

In Greece, MMR vaccine was introduced in 1980 for boys and girls at 1 yearwith no policy relating to targets for vaccine (Panagiotopoulos et al., 1999). Thisalmost certainly resulted in MMR being given mainly to children in the privatesector with population coverage consistently below 50% and increased rates ofsusceptibility in older groups. In 1993, rubella incidence in young adults was higherthan in any previous epidemic year. Twenty-five children were diagnosed withdefects associated with congenital rubella infections, 24.6 per 100,000 live births,amounting to the largest epidemic since 1950. With the gradual increase in theproportion of susceptible women of childbearing age a subsequent outbreak oc-curred in the winter of 1998/1999 (Panagiotopoulos and Georgakopoulou, 2004),which resulted in around 2500 domestic cases (Giannakos et al., 2000). It also led torubella being imported to the UK by Greek students attending British universitiesand led to the only congenital rubella birth reported in the UK in 1999 (Tookey etal., 2000).

Schoolgirl vaccination was introduced in Australia in 1971. This resulted in asignificant decline in congenital rubella but cases continued to occur, even in thosevaccinated, as a result of antibody loss or poor vaccine response. In view of thispostpartum vaccination was introduced for women with low-level rubella anti-bodies detected at antenatal testing (Cheffins et al., 1998). In 1989, MMR vaccinewas introduced for infants at 1 year and the adolescent programme for girls wasdiscontinued in 1994–1995. In 1998 a second dose of MMR was recommended at4 years. National surveillance systems were in place for the reporting of clinical andcongenital rubella, which showed that in 1992–1995 the risk of rubella was >20/100,000 and declined to 7.2 per 100,000 in 1997 (Sullivan et al., 1999).


In Finland, rubella immunisation for 11–13-year-old girls was introduced in1975 with seroprevalence in women increasing to a plateau of 90% by 17 years ofage. Following the introduction of MMR vaccination for young children in 1982seroprevalence among children aged 2–3 years of age had increased to 95% by1986. High vaccine coverage of over 97% has been sustained and rubella virtuallyeliminated with no cases of congenital rubella reported since 1986 (Davidkin et al.,2004).

In Sri Lanka (Gunasekera and Gunasekera, 1996), an epidemic of rubella oc-curred in 1994–1995 with 275 cases of congenital rubella reported in 1994 and 169in 1995. Although rubella vaccine was then widely available in the private sector itwas not available to the general population. Subsequently a national programmewas introduced for all 11–15-year-old girls and, on request, for susceptible womenaged 16–44 years.

In the USA, a different approach was taken and a policy of universal rubellaimmunisation was adopted in 1969. Rubella vaccine was given to boys and girls intheir second year of life with the aim of developing herd immunity so that womenwho were susceptible to rubella would not be exposed, particularly to infectionfrom their own children who posed the greatest risk. This approach had a dramaticeffect and the wide use of rubella vaccine reduced the incidence of rubella andcongenital rubella by >99% from pre-vaccine levels (Reef et al., 2000). In 1995,State Health Departments reported only 138 cases of rubella and 6 cases of CRS toCDC, demonstrating that the goal to eliminate indigenous congenital rubella by2000 was an achievable target (Reef et al., 2002). In the US, rubella is nearingelimination but sporadic outbreaks, mainly among unvaccinated foreign residents,still pose challenges for its eradication (MMWR Morbidity Mortality WeeklyReport, 2005).

Following the implementation of measles vaccination programmes in theAmericas, regional measles surveillance systems documented widespread rubellavirus circulating in many countries with a potential for major rubella epidemics(Castillo-Solorzano et al., 2003). As a result rubella prevention efforts werestrengthened and the goal of rubella elimination added to the current measlesimmunisation programme. Countries in the Americas are at different stages ofrubella prevention strategies but the majority now recommend the inclusion ofMMR vaccine in their routine childhood programmes. In order to accelerate thecongenital rubella prevention programme and reduce the number of women ofchildbearing age who are susceptible to rubella, many countries undertook one-offmass vaccination campaigns targeting females aged 5–39 years with measles andrubella-containing vaccines. In some countries both males and females were tar-geted to reduce the circulation of virus in the community and interrupt virustransmission. In order to assess the burden of rubella disease, integrated surveil-lance systems were developed for measles and rubella as well as for congenitalrubella and laboratory reports of virus isolation (Castillo-Solorzano et al., 2003).

More recently in Brazil, and Costa Rica and the West Indies, mass immuni-sation programmes have been introduced to avoid outbreaks of measles and


rubella. The impact of these combined rubella vaccination strategies and the rapidinterruption of rubella virus transmission is already evident in Cuba, Brazil, CostaRica, Chile and the English-speaking Caribbean (Massad et al., 1995; Castillo-Solorzano et al., 2003).

In 1999, there was a large rubella epidemic in Costa Rica with the highestattack rate in adults 20–39 years and around 500 cases occurring in pregnantwomen. This was likely to have resulted from the shift in age-specific incidence ofinfection resulting from the introduction of MMR vaccine in children. In responseto this outbreak, women aged 18–40 were offered MMR vaccine in a campaigncarried out in two phases: non-pregnant women and then pregnant women.Women were informed about the safety of the vaccine and pregnancy outcomesamong women vaccinated in pregnancy were monitored.

A similar situation arose in Brazil where women of childbearing age were vac-cinated in a mass campaign and those who were subsequently found to be pregnantare being systematically followed up and pregnancy outcome assessed (EPI News-letter, 2002). From the information gathered in these investigations it should bepossible to assess the safety of giving rubella vaccine to pregnant women. However,it will be essential to follow up the infants exposed to rubella vaccine in utero todetermine their infection status and for those infected longer term follow-up will berequired to identify defects that may not be apparent at birth.

In 1996 a survey on worldwide rubella vaccination use, sponsored by WHO,identified 78/214 (36%) countries with a national policy for rubella immunisationand inclusion of rubella vaccine in their national immunisation programmes. Sev-enty-eight countries had a national policy of rubella vaccination. This included92% of industrialised countries and 12% of developing countries (Robertson et al.,1997). By 2004, this had increased to 116 (54%) of the 214 countries and territoriesreporting to the WHO (World Health Organisation, 2006).

Surveillance of rubella and congenital rubella

In order to monitor the efficacy of a rubella immunisation programme and todetect unforeseen problems that might arise, surveillance systems need to be inplace. WHO developed guidelines for the surveillance of rubella and congenitalrubella in 1999 (Cutts et al., 1999), which were updated in 2003 (World HealthOrganisation, 2003). These recommend clinical surveillance of rubella and con-genital rubella infection; serological surveillance to detect changes in the prevalenceof rubella immunity in the population; and the recording of vaccine coverage. Theage- and sex-specific incidence of rubella infection and the incidence of congenitalrubella infection should be made available at the country level.

As an example of national rubella surveillance, and to demonstrate the differentcomponents that need to be in place, we describe below the national surveillance ofcongenital rubella in the UK. Comparable systems are in operation in Canada,Australia, the US and other countries. Mass vaccination programmes change theepidemiological dynamics of rubella infection within the population and potential


long-term effects, such as an increase in rubella susceptibility among women ofchildbearing age, should be anticipated by surveillance. Thus, a vaccination policycan be appropriately adapted to avoid complications.

Surveillance of clinical disease and seroprevalence

Laboratory confirmed cases of rubella are notified to the Health Protection AgencyCentre for Infections. However, clinical and laboratory surveillance of rubella in-fection is of limited value as most infections are subclinical and will not bereported. Serological surveys are therefore used to monitor the prevalence ofrubella antibodies in women of childbearing age in different sections of thecommunity. These are usually based on antenatal sera, reflecting the seroprevalenceof rubella-specific antibody in pregnant women, and provide information for policydecisions.

Surveillance of congenital rubella

The National Congenital Rubella Surveillance Programme (NCRSP) was estab-lished in 1971 to monitor the effect of the rubella vaccination policy in England,Scotland and Wales. Passive reporting through laboratory reporting systems andfrom audiologists and paediatricians was supplemented in 1990 by an active pae-diatric reporting scheme run through the British Paediatric Surveillance Unit of theRoyal College of Paediatrics and Child Health (Nicoll et al., 2000). Data on ma-ternal demographic characteristics and infection and vaccination history, as well asthe clinical presentation in the infant and laboratory results, are routinely collectedin order to monitor the risk factors associated with congenital rubella cases and toprovide information on the changing epidemiology of congenital rubella in an eraof prolonged high vaccination uptake.

The NCRSP data combined with routinely collected statistics from the Officefor National Statistics on rubella-associated terminations of pregnancy demon-strate the dramatic decline in the number of pregnancies affected by rubella in-fection (see Fig. 1), from an average of about 40 births and 400 terminationsreported each year in the period 1976–1980 to an average of about 4 births and 8terminations a year in the 1990s. Data on rubella-associated terminations are nolonger published since the annual number is so low (o10). About a third of birthsreported since 1990 were to women who acquired their infection in early pregnancyoutside the UK, and another third were to women who had arrived relativelyrecently in the UK and originated from countries without well-established immu-nisation programmes (Tookey, 2004).

This surveillance system has proved sensitive to small changes in incidence ofcongenital rubella. For example, in 1996 following a small outbreak of infectionmainly among students, most cases were reported within 2 months of birth. Linkswith outbreaks elsewhere can also be observed, for example the one case reportedin 1999 was associated with an outbreak in Greece (Tookey et al., 2000). However,


since the introduction of MMR, most of the cases reported have had multipleserious defects and it is likely that there has been an under ascertainment of singledefects such as congenital rubella deafness. Deafness may not be detected untilafter the child receives MMR and the presence of vaccine immunity cannot easilybe distinguished from the persistence of immunity resulting from congenital infec-tion. Furthermore, as congenital rubella is so rare, it is unlikely to be considered asa potential explanation for isolated hearing loss, or other non-specific symptoms.As a result, in contrast to the period prior to the introduction of MMR vaccine,most reported cases are now severely symptomatic at birth. Between 1999 and2005, 13 infants with congenital rubella were reported including one born in Ire-land, and one stillborn infant. Nine were imported cases with maternal infectionacquired abroad, and four were born to women who acquired the infection in theUK. Only two affected children had mothers who were born in the UK.

Conclusion

The introduction of rubella vaccine programmes has had a marked impact on theprevention of congenital rubella in developed countries, although cases still con-tinue to occur. Few cases of rubella infection are now reported in the US or theUK, where surveillance systems are in place, and the few babies reported to havecongenital infection are usually born to women from countries with no rubella

0

10

20

30

40

50

60

70

80

90

71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 '01 '03

Year of birth

Bir

ths

0

200

400

600

800

1000

1200

Te

rm

ina

tio

ns

Congenital Rubella births

(England, Scotland, Wales)

Terminations (rubella disease/

contact) (England & Wales only)

Fig. 1 Notifications of congenital rubella births in England, Scotland and Wales and notifications of

rubella-associated terminations of pregnancy in England and Wales 1971–2005.


vaccination programme or poor vaccine uptake rates (Reef et al., 2002; Tookey,2004).

High rubella vaccine uptake must be maintained if rubella is to be eliminated.Furthermore, it is essential to have good surveillance systems in place so thatchanges in rubella susceptibility can be monitored and steps taken if susceptiblegroups start to emerge. There is also a need for greater public understanding of whyvaccination programmes are important and the implications of low uptake.

The global burden of rubella and congenital rubella remains undefined in manydeveloping countries. However, there is a risk that the incidence of rubella inwomen of childbearing age will increase in the absence of high vaccine coveragewith a potential risk of adverse pregnancy outcomes and rubella-associated dis-ability in surviving children. High levels of vaccine uptake need to be maintained ifcongenital rubella is to be eradicated. The European region of WHO have set atarget for rubella elimination in the 52 countries of the region by 2010 (WeeklyEpidemiological Record, 2005). There is still some way to go but until rubella iseradicated, unvaccinated individuals or communities will remain susceptible toimported infection, which could promote indigenous transmission. Even in coun-tries with high vaccine coverage there is a risk of reintroduction of infection fromneighbouring countries where vaccine uptake is not maintained. This is particularlypertinent for Europe where there is frequent travel across country borders. Theheightened risk of rubella in immigrants from countries where vaccination is notuniversal has been highlighted in surveillance systems and reinforces the need forvaccination programmes to take this into account. Virus genotyping to identify thegeographical origin of the infection is also a useful resource in surveillance pro-grammes, particularly in countries such as the US and the UK where most cases ofcongenital rubella are due to imported infections.

More research is also needed regarding waning immunity following vaccina-tion, particularly for the large cohorts of individuals in developed countries whohave received their rubella vaccine at a very young age as part of an MMRimmunisation programme. These countries have little circulating wild virus, andtherefore no natural triggers for prolonged immunity. It is possible that rubellavaccine boosters are needed throughout life to prevent outbreaks of rubella inpregnant women in countries that have opted for the universal model of rubellavaccination.

References

Anderson RM, Grenfell BT. Quantitative investigations of different vaccination policies for

the control of congenital rubella syndrome (CRS) in the United Kingdom. J Hyg (Lond)

1986; 96: 305–333.

Anderson RM, May RM. Vaccination against rubella and measles: quantitative investiga-

tions of different policies. J Hyg (Lond) 1983; 90: 259–325.


Anderson RM, May RM. Infectious Diseases of Humans: Dynamics and Control. Oxford:

Oxford University Press; 1991.

Best J, Banatvala JE. Rubella. In: Principles and Practice of Clinical Virology (Zuckerman

AJ, Banatvala JE, Pattison JR, Griffiths PD, Schoub BD, editors). Chichester: Wiley;

2004.

Briss PA, Fehrs LJ, Hutcheson RH, Schaffner W. Rubella among the Amish: resurgent

disease in a highly susceptible community. Pediatr Infect Dis J 1992; 11: 955–959.

Castillo-Solorzano C, Carrasco P, Tambini G, Reef S, Brana M, de Quadros CA. New

horizons in the control of rubella and prevention of congenital rubella syndrome in the

Americas. J Infect Dis 2003; 187(Suppl 1): S146–S152.

Celers J, Aymard M, Feingold J, Godde-Jolly D. [Rubella and congenital cataracts]. Arch Fr

Pediatr 1983; 40: 391–395.

Cheffins T, Chan A, Keane RJ, Haan EA, Hall R. The impact of rubella immunisation on

the incidence of rubella, congenital rubella syndrome and rubella-related terminations of

pregnancy in South Australia. Br J Obstet Gynaecol 1998; 105: 998–1004.

Christenson B, Bottiger M. Long-term follow-up study of rubella antibodies in naturally

immune and vaccinated young adults. Vaccine 1994; 12: 41–45.

Cooper LZ, Alford CA. Rubella. In: Infectious Diseases of the Fetus and the Newborn

Infant (Remington JS, Klein JO, editors). Philadelphia: W.B. Saunders Company; 2001;

pp. 347–388.

Cutts FT, Abebe A, Messele T, Dejene A, Enquselassie F, Nigatu W, et al. Sero-epidemio-

logy of rubella in the urban population of Addis Ababa, Ethiopia. Epidemiol Infect 2000;

124: 467–479.

Cutts FT, Best J, Siqueira MM, Engstrom K, Robertson SE. Guidelines for Surveillance of

Congenital Rubella Syndrome (CRS) and Rubella. Geneva: World Health Organisation,

Department of Vaccines and Other Biologicals; 1999.

Cutts FT, Robertson SE, Diaz-Ortega JL, Samuel R. Control of rubella and congenital

rubella syndrome (CRS) in developing countries, Part 1: burden of disease from CRS. Bull

World Health Organ 1997; 75: 55–68.

Cutts FT, Vynnycky E. Modelling the incidence of congenital rubella syndrome in deve-

loping countries. Int J Epidemiol 1999; 28: 1176–1184.

Davidkin I, Peltola H, Leinikki P. Epidemiology of rubella in Finland. Euro Surveill 2004; 9:

13–14.

Davidkin I, Peltola H, Leinikki P, Valle M. Duration of rubella immunity induced by two-

dose measles, mumps and rubella (MMR) vaccination. A 15-year follow-up in Finland.

Vaccine 2000; 18: 3106–3112.

de Haas R, van den HS, Berbers GA, de Melker HE, Conyn-van Spaendonck MA. Preva-

lence of antibodies against rubella virus in the Netherlands 9 years after changing from

selective to mass vaccination. Epidemiol Infect 1999; 123: 263–270.

de Melker HE, van den HS, Berbers GA, Conyn-van Spaendonck MA. Evaluation of the

national immunisation programme in the Netherlands: immunity to diphtheria, tetanus,

poliomyelitis, measles, mumps, rubella and Haemophilus influenzae type b. Vaccine 2003;

21: 716–720.

Eckstein MB, Brown DW, Foster A, Richards AF, Gilbert CE, Vijayalakshmi P. Congenital

rubella in south India: diagnosis using saliva from infants with cataract. BMJ 1996; 312: 161.

Enders G, Nickerl-Pacher U, Miller E, Cradock-Watson JE. Outcome of confirmed peri-

conceptional maternal rubella. Lancet 1988; 1: 1445–1447.


EPI Newsletter. Brazil accelerates control of rubella and prevention of congenital rubella

syndrome. EPI Newsl 2002; 24: 1–3.

Forrest JM, Menser MA, Burgess JA. High frequency of diabetes mellitus in young adults

with congenital rubella. Lancet 1971; 2: 332–334.

Forrest JM, Turnbull FM, Sholler GF, Hawker RE, Martin FJ, Doran TT, et al. Gregg’s

congenital rubella patients 60 years later. Med J Aust 2002; 177: 664–667.

Giannakos G, Pirounaki M, Hadjichristodoulou C. Incidence of congenital rubella in Greece

has decreased. BMJ 2000; 320: 1408.

Grillner L, Forsgren M, Barr B, Bottiger M, Danielsson L, De Verdier C. Outcome of

rubella during pregnancy with special reference to the 17th–24th weeks of gestation. Scand

J Infect Dis 1983; 15: 321–325.

Gunasekera DP, Gunasekera PC. Rubella immunisation-learning from developed countries.

Lancet 1996; 347: 1694–1695.

Gupta RK, Best J, MacMahon E. Mumps and the UK epidemic 2005. BMJ 2005; 330:

1132–1135.

Hanratty B, Holt T, Duffell E, Patterson W, Ramsay M, White JM, et al. UK measles

outbreak in non-immune anthroposophic communities: the implications for the elimina-

tion of measles from Europe. Epidemiol Infect 2000; 125: 377–383.

Health Protection Agency. Increase in measles cases in 2006, in England and Wales. CDR

Wkly 2006; 16: 3.

HMSO. Immunisation against Infectious Disease. London: HMSO, Department of Health;

1996.

Hofmann J, Kortung M, Pustowoit B, Faber R, Piskazeck U, Liebert UG. Persistent fetal

rubella vaccine virus infection following inadvertent vaccination during early pregnancy.

J Med Virol 2000; 61: 155–158.

Jansen VA, Stollenwerk N, Jensen HJ, Ramsay ME, Edmunds WJ, Rhodes CJ. Measles

outbreaks in a population with declining vaccine uptake. Science 2003; 301: 804.

Knox EG. Theoretical aspects of rubella vaccination strategies. Rev Infect Dis 1985; 7(Suppl

1): S194–S197.

Marshall WC. The clinical impact of intrauterine rubella. Ciba Found Symp 1972; 10: 3–22.

Massad E, Azevedo-Neto RS, Burattini MN, Zanetta DM, Coutinho FA, Yang HM, et al.

Assessing the efficacy of a mixed vaccination strategy against rubella in Sao Paulo, Brazil.

Int J Epidemiol 1995; 24: 842–850.

McBrien J, Murphy J, Gill D, Cronin M, O’Donovan C, Cafferkey MT. Measles outbreak in

Dublin, 2000. Pediatr Infect Dis J 2003; 22: 580–584.

McIntosh ED, Menser MA. A fifty-year follow-up of congenital rubella. Lancet 1992; 340:

414–415.

Mellinger AK, Cragan JD, Atkinson WL, Williams WW, Kleger B, Kimber RG, et al. High

incidence of congenital rubella syndrome after a rubella outbreak. Pediatr Infect Dis J

1995; 14: 573–578.

Menser MA, Dods L, Harley JD. A twenty-five-year follow-up of congenital rubella. Lancet

1967; 2: 1347–1350.

Menser MA, Forrest JM, Bransby RD. Rubella infection and diabetes mellitus. Lancet 1978;

1: 57–60.

Miller CL. Rubella in the developing world. Epidemiol Infect 1991; 107: 63–68.

Miller E. Rubella reinfection. Arch. Dis Child 1990; 65: 820–821.

Miller E. Rubella in the United Kingdom. Epidemiol Infect 1991; 107: 31–42.



successive stages of pregnancy. Lancet 1982; 2: 781–784.

Miller E, Waight P, Gay N, Ramsay M, Vurdien J, Morgan-Capner P, et al. The epide-

miology of rubella in England and Wales before and after the 1994 measles and rubella

vaccination campaign: fourth joint report from the PHLS and the National Congenital

Rubella Surveillance Programme. Commun Dis Rep CDR Rev 1997; 7: R26–R32.

Miller E, Waight PA, Vurdien JE, White JM, Jones G, Miller BH, et al. Rubella surveillance

to December 1990: a joint report from the PHLS and National Congenital Rubella Sur-

veillance Programme. CDR (Lond Engl Rev) 1991; 1: R33–R37.

MMWR Morbidity Mortality Weekly Report. Elimination of rubella and congenital rubella

syndrome—United States, 1969–2004. MMWR Morb Mortal Wkly Rep 2005; 54:

279–282.

Morgan-Capner P, Miller E, Vurdien JE, Ramsay ME. Outcome of pregnancy after ma-

ternal reinfection with rubella. CDR (Lond Engl Rev) 1991; 1: R57–R59.

NHS Health and Social Care Information Centre. NHS Immunisation Statistics, England:

2004–05, Department of Health; 2005.

Nicoll A, Lynn R, Rahi J, Verity C, Haines L. Public health outputs from the British

Paediatric Surveillance Unit and similar clinician-based systems. J R Soc Med 2000; 93:

580–585.

Nokes DJ, Anderson RM. Measles, mumps, and rubella vaccine: what coverage to block

transmission? Lancet 1988; 2: 1374.

Panagiotopoulos T, Antoniadou I, Valassi-Adam E. Increase in congenital rubella occur-

rence after immunisation in Greece: retrospective survey and systematic review. BMJ

1999; 319: 1462–1467.

Panagiotopoulos T, Georgakopoulou T. Epidemiology of rubella and congenital rubella

syndrome in Greece, 1994–2003. Euro Surveill 2004; 9: 17–19.

Peckham C. Congenital rubella in the United Kingdom before 1970: the prevaccine era. Rev

Infect Dis 1985; 7(Suppl 1): S11–S16.

Peckham CS. Clinical and laboratory study of children exposed in utero to maternal rubella.

Arch Dis Child 1972; 47: 571–577.

Peckham CS, Martin JA, Marshall WC, Dudgeon JA. Congenital rubella deafness: a pre-

ventable disease. Lancet 1979; 1: 258–261.

Rafila A, Marin M, Pistol A, Nicolaiciuc D, Lupulescu E, Uzicanin A, et al. A large rubella

outbreak, Romania—2003. Euro Surveill 2004; 9: 7–9.

Reef SE, Frey TK, Theall K, Abernathy E, Burnett CL, Icenogle J, et al. The changing

epidemiology of rubella in the 1990s: on the verge of elimination and new challenges for

control and prevention. JAMA 2002; 287: 464–472.

Reef SE, Plotkin S, Cordero JF, Katz M, Cooper L, Schwartz B, et al. Preparing for

elimination of congenital Rubella syndrome (CRS): summary of a workshop on CRS

elimination in the United States. Clin Infect Dis 2000; 31: 85–95.

Robertson SE, Cutts FT, Samuel R, Diaz-Ortega JL. Control of rubella and congenital

rubella syndrome (CRS) in developing countries, Part 2: vaccination against rubella. Bull

World Health Organ 1997; 75: 69–80.

Sullivan EM, Burgess MA, Forrest JM. The epidemiology of rubella and congenital rubella

in Australia, 1992 to 1997. Commun Dis Intell 1999; 23: 209–214.

Tookey P. Pregnancy is contraindication for rubella vaccination still. BMJ 2001; 322: 1489.

Tookey P. Rubella in England, Scotland and Wales. Euro Surveill 2004; 9: 21–22.


Tookey P, Molyneaux P, Helms P. UK case of congenital rubella can be linked to Greek

cases. BMJ 2000; 321: 766–767.

Tookey PA, Cortina-Borja M, Peckham CS. Rubella susceptibility among pregnant women

in North London, 1996–1999. J Public Health Med 2002; 24: 211–216.

Tookey PA, Peckham CS. Surveillance of congenital rubella in Great Britain, 1971–96. BMJ

1999; 318: 769–770.

Ushida M, Katow S, Furukawa S. Congenital rubella syndrome due to infection after

maternal antibody conversion with vaccine. Jpn J Infect Dis 2003; 56: 68–69.

van der Veen Y, Hahne S, Ruijs H, van Binnendijk R, Timen A, van Loon A, et al. Rubella

outbreak in an unvaccinated religious community in the Netherlands leads to cases of

congenital rubella syndrome. Euro Surveill 2005; 10: 274–275.

Vynnycky E, Gay NJ, Cutts FT. The predicted impact of private sector MMR vaccination

on the burden of congenital rubella syndrome. Vaccine 2003; 21: 2708–2719.

Weekly Epidemiological Record. Progress towards elimination of measles and prevention of

congenital rubella infection in the WHO European Region, 1990–2004. Wkly Epidemiol

Rec 2005; 80: 66–71.

World Health Organisation, Department of Vaccines and Biologicals. Report of a Meeting

on Preventing Congenital Rubella Syndrome: Immunization Strategies, Surveillance

Needs. Geneva: World Health Organisation; 2000.

World Health Organisation. Surveillance Guidelines for Measles and Congenital Rubella

Infection in the WHO European Region. Copenhagen: WHO, Europe; 2003.

World Health Organisation. Rubella and Congenital Rubella Syndrome. http://www.

who.int/immunization_monitoring/diseases/rubella/en/ . Accessed 8th May 2006.

(Electronic Citation).

Zgorniak-Nowosielska I, Zawilinska B, Szostek S. Rubella infection during pregnancy in the

1985–86 epidemic: follow-up after seven years. Eur J Epidemiol 1996; 12: 303–308.


Future Requirements

Successful rubella vaccination programmes have resulted in the virtual elimination ofpostnatally and congenitally acquired rubella in many countries (see Chapter 4 & 5).However, countries wishing to eliminate rubella must not only ensure high vaccineuptake levels, but programmes should be supported by high-quality surveillance,including molecular epidemiological studies, in order to obtain evidence about thecirculation of indigenous viruses, or the importation of new strains from other partsof the world. This is particularly relevant when monitoring the later stages of rubellaelimination.

Surveillance is dependent on not only employing techniques appropriate tolocal health services, but clinical and laboratory investigations must be monitoredindependently by those with appropriate experience.

There are still a number of countries, particularly in Sub Saharan Africa, inwhich there are no programmes for rubella elimination. However, it is importantthat such countries carry out epidemiological studies in order to determine theprevalence of rubella antibodies in different populations, and assess, if possible, theimpact of congenitally acquired rubella on health services.

The development of robust laboratory investigations, which can be used inthe field at ambient temperatures in developing countries, should be a priority,particularly for rubella-specific IgG and IgM, e.g. as has been accomplished forDengue.

As infectious diseases become increasingly rare as a result of successful immu-nisation programmes, there is a tendency for more attention to be focused by themedia on rare or even unrelated vaccine side effects. Unfortunately, unfounded andsensationalist reporting by the media may discourage vaccine uptake, as has beenthe case with MMR in the UK. This has resulted in a marked decline in vaccineuptake rates in recent years. WHO’s Vaccination Safety Advisory Committee andnational bodies, e.g. in the UK the Department of Health, The Medical ResearchCouncil and The Health Protection Agency, must be prepared to provide reliable,independent and scientific assessment on vaccine safety issues, preparing reportsand liaising with the media within as short a time as possible.

Some of those involved in vaccination programmes continue to express concernabout long-term protection among those with low or no longer detectable serumantibody levels some years after vaccination, particularly in countries whereexposure to naturally acquired rubella becomes increasingly uncommon. Although

115

challenge studies (see chapter 4) suggest that such persons are immune, researchdirected towards determining B- and T-cell immunity would be of value since thismay result in eliminating the unnecessary expense of giving booster doses of vac-cine to those with low or no longer detectable antibody responses.

Although current rubella vaccines are safe, well tolerated and immunogenic,this should not be a deterrent from developing new vaccines; aerosolised prepa-rations show promise and obviate the need to use needles with the attendant risksof transmitting blood-borne infections in developing countries (see chapter 4).

Future Requirements116

List of Contributors

Jangu E. BanatvalaKing’s College London MedicalSchool at Guy’s, King’s College andSt Thomas’ HospitalsLondon, SEI 7EH, UK

Jennifer M. BestKing’s College London School ofMedicine at Guy’s, King’s College andSt Thomas’ HospitalsLondon, SEI 7EH, UK

Min-Hsin ChenDivision of Viral DiseasesCenters for Disease Control and Prevention1600 Clifton RoadAtlanta, GA 30333, USA

Gisela EndersInstitute of VirologyInfectiology and Epidemiology e.V and LaborProf. Dr. med Gisela Enders and PartnerRosenbergstraXe 8570193, Stuttgart, Germany

Pia HardelidCentre for Paediatric Epidemiology and BiostatisticsUCL Institute of Child Health30 Guilford StreetLondon, WCIN 1EH, UK

Joseph IcenogleDivision of Viral DiseasesCenters for Disease Control and Prevention1600 Clifton RoadAtlanta, GA 30333, USA

117

Catherine PeckhamCentre for Paediatric Epidemiology and BiostatisticsUCL Institute of Child Health30 Guilford StreetLondon, WCIN 1EH, UK

Stanley A. PlotkinUniversity of Pennsylvania and Sanofi Pasteur4650 Wismer RoadDoylestown, PA 18901, USA

Susan ReefRubella and Mumps ActivityNational Immunization ProgramCenters for Disease Control and Prevention1600 Clifton RoadAtlanta, GA 30333, USA

Pat TookeyCentre for Paediatric Epidemiology and BiostatisticsUCL Institute of Child Health30 Guilford StreetLondon, WCIN 1EH, UK

List of Contributors118

Index

Page numbers suffixed by t and f refer to Tables and Figures respectively.

30 terminal stem loop 7

50 cap structure 6

50 stem-loop 7

abortion, therapeutic 97

adenoviruses 21

aerosol

person-person transmission 19

vaccine administration 81

alphavirus

conformational changes 7

genome 8

genus 1, 12

receptor binding sites 5

sequence homology to rubella virus 1

structure 2

transmission 1

amniocentesis 60

amniotic fluid 60, 62–63, 101

antenatal screening 53

antibodies

HAI 40, 68

hyperimmune 51

monoclonal 32, 42–43, 51

neutralizing 40, 42, 82, 84

pre-existing 57

protective 42, 68

antigen 26, 41

recombinant 40, 43, 46

antigen-antibody complexes 20

arboviruses 21

arteries, damage of 30

arthralgia 21, 84–85

arthritis 21, 84–85

arthropathy 84–85

autism 25, 32

autoimmune 25, 32, 98

B-cell(s)

epitopes 32, 40

immunity 115

behavioural disorders 28t, 32

Benoit strains 83

BHK cells 12, 40

blindness 31

British Paediatric Surveillance Unit 108

budding 2, 11

capsid (C) protein

antibodies to 46, 51

in nucleoclapid formation 2

role in virus entry 7

role in replication 7, 10–11

phosphorylation state of 11

cardiac anomalies x

cardiac defects 95

cardiovascular anomalies 29

carpel tunnel syndrome 23

cataracts x, 25, t28, 30–31, 66, 79, 95–96

cell death 10

cell-mediated immunity 26

cellular

immune responses 82

proliferation 24

retardation 24

survival 11

Center for Disease Control (CDC), US 106

Chikungunya virus 21, 54

chorionic villae 24

biopsies of 60, 62

chronic inflammatory joint disease 21

cis-acting elements 7–8

citron-K-kinase 11

clathrin-mediated endocytosis 7

clinical diagnosis 21

congenital anomalies x, 25, 27

congenital malformation x

congenital rubella

asymptomatic 64

diagnosis of 40, 50, 64t, 69

retrospective 96, 98

eradication of 103

late-onset manifestations of 32, 98

prevalence of 96

prevention of 98

surveillance of 98, 107–108

congenital rubella syndrome (CRS) 1, 24, 96

cases of 14, 79

clinical features of xi, 27–28, 28t

diagnosis of 54–55, 66

119

due to maternal infection 23–27

due to maternal re-infection 84

due to vaccination 86–87

elimination of 12

long-term prognosis of 33

treatment costs 89

cord blood 101

cordocentesis 60

corneal

abnormalities 31

hydrops 31

opacities 31

corticosteroids 29

cost-benefit analysis (CBA) 89

cost-effectiveness analysis (CEA) 89

countries

developing 107, 110

industrialized 98, 107, 110

CNS complications 23, 25

CSF

rubella antibody titres 32

rubella-specific IgG

cytomegalovirus 43

cytopathic effect (CPE) 10, 39, 42, 51f

cytopathic vacuoles 3f, 7, 11

deafness x, 25, 27, 30, 66, 69, 97, 109

defects

congenital 28, 28t, 57, 97, 101

eye 30–31

hearing 27, 30, 97

septal 29

demylination 23, 32

dengue virus 54

Department of Health, UK 115

diabetes mellitus 28, 98 see also Type 1 and Type 2

diagnosis

false-positive results 41, 43, 63

false-negative results 43, 52, 61, 63

diarrhoea 29

diethylamine (DEA) 44

shift value method (DSV) 44, 67

wash method 44, 50

dried blood spots (DBS)

dysgammaglobulinaemia 25

E1 glycoprotein

antibodies to 46, 51

antigenicity 2

epitopes 40, 68

conformational change 7

cytoplasmic tail 2

fusion peptide 4, 7

glycosylation of 2, 4, 11

immune tolerance to 26

processing of 11

signal peptide 5

E2 glycoprotein

antibodies to 46–47, 51

conformational change 7

epitopes 68

glycosylation of 4, 11

processing of 11

signal peptide 2

egg allergies 86

emboli 24

embyropathy x, 24

encephalitis 31, 98

endemic 98

endocrine disorders 32

endoplasmic reticulum 3, 11

enteroviruses 21, 54

enzyme immunoassay (EIA) 40–42, 53, 63, 66,

68, 82

enzyme-linked immunospot test (ELISPOT) 67

epidemic(s) 25, 80, 98, 100, 103, 105–107

epidemiology 12, 13, 87, 95–96, 104

epitope(s) 4, 32, 40, 68

Epstein-Barr virus (EBV) 43

Fallot’s tetrology 30

fetal

blood 60, 62–63

circulation 24

damage 23, 24, 57, 60, 97

infection 25, 27, 56, 95

immune responses 24, 27

immunoglobulin 24

organs 24

fever 1, 20, 84

first trimester x, 23–27, 32, 58, 95, 97

genotype 12

German measles ix

gestation 24, 26, 55, 60, 61, 97

gestational age, role of 26f

glaucoma 28t, 31, 98

glycoprotein

E1 2

E2 2

E1-E2 heterodimer 2

glycosylation

N-linked 2, 4

O-linked 4

120

Golgi 11–12, 51

group A streptococcus 49, 54

growth hormone deficiency 33, 98

growth retardation 28t, 33, 95

Gullain-Barre syndrome 23

Guthrie cards 50

haemagglutination inhibition 4, 80

haemagglutination inhibition (HAI) test 39,

41–42, 53–54, 57, 68, 80

haematological abnormalities 23

haemolytic anaemia 23

hairpin structure 8

Hashimoto’s disease 25

Health Protection Agency, UK 108,

115

hearing impairment 95, 97

helicase 9–10

herd immunity 99, 102, 106

histological changes 24

HIV 23, 86

human diploid cells 80

human fibroblast cells 86

human herpesvirus-6 (HHV-6) 49, 54

human herpesvirus-7 (HHV-7) 54

hypertension 30

hyperthyroidism 33, 98

hypothyroidism 33, 98

indirect immunofluorescence 51

IgG antibodies, rubella specific

as part of immune response 82

detection of 50

maternal 24

removal of 42

screening of 53–55

tests for 41–42

IgG avidity tests 43–45, 57, 67

IgM antibodies, rubella-specific

as part of immune response 82–83

detection of 42–43, 50

long-persisting 45, 58

production of 61, 101

screening of 53–55

tests for 63

IgM capture assay 42–43, 63, 69

immigrants 100, 110

immigration patterns 99

immune complexes 21, 29, 44

immune response

cell-mediated 84

cellular 82

fetal 24

humoral 20

immunity x

cell-mediated 23, 26

naturally acquired 104

persistence of 109

prevalence of 107

testing 42

vaccine-acquired 56, 68, 97, 104, 109

waning 104–105

immunization 99, 104

campaign 83

programmes 87–89, 106–108

immunoblotting (IB) 40, 57

immunocompromised 23

immunodeficiency diseases 86

immunoglobulins 25

immunological memory 68

immunosupression 23, 86

in vitro fertilization 53

in utero x, 1, 26, 32, 107

infection

asymptomatic 63, 95–96, 109

congenital 25, 27, 32, 68, 98, 101,

105

fetal, risk of 97

imported 110

intrauterine 27

maternal 24–27, 63, 87, 96, 109

persistent x, 1, 23

primary 45, 47–48

subclinical 20, 23, 108

systemic 19

transmission of 19

infectious cDNA clone 9

insulin-dependent diabetes mellitus (IDDM)

32–33

interferon synthesis 26

intracranial calcification 31, 60

intragenic region 9

joint symptoms

due to rubella infection 21

due to rubella vaccination 84–85

in relation to menstrual cycle 21

keratoconus 31

last menstrual period (LMP) 27, 55, 97,

101

latex agglutination test (LA) 41, 53

121

lesions

bone 28, 28t

macular 20

vascular 31

lethargy 31

leukaemia 86

lymphoblastic 23

linear hyperechogenicity 31

lymphadenopathy 19, 28t, 84

lymphocytic pleocytosis 23

lymph nodes 20

lymphokines 24

lymphoproliferative assay 67

lymphoproliferative responses 26, 68

maternal viraemia 24

measles ix, 19, 23, 49, 52, 54

cross-reacting antibodies 43

molecular epidemiology of 12

outbreaks of 103

vaccine 80–82

Edmonston-Zagreb (EZ) 81

Schwartz 81

vaccination programs 106

menopause 33

mental retardation 95

methyltransferase 9

microcephaly 28t, 31

microphthalmia x, 28t, 30–31

mitochondrial matrix protein 10

MR vaccines 81

MMR

vaccine 23, 49, 69, 81–82, 101–103, 105, 107, 109

vaccination 55, 66, 86, 101–103, 105, 110

molecular mimicry 32

multi-plex PCR (mPCR) 52

multi-system disease 23

mortality, perinatal 30

mumps vaccine 81–82

mutagenesis 9

myocarditis t28, 30

nasopharyngeal

secretions 21, 25–26, 64

swabs 64

National Congenital Rubella Surveillance

Programme (NCRSP), UK 87, 108

necrosis 24

neonatal diagnosis 63

nested reverse transcription polymerase chain

reaction (RT-nPCR) 40, 52, 60–64

neuropathy 80, 84

neutralization tests (NT) 42

non-reducing immunoblot (IB) assay 45–46,

66

non-structural proteins (NSPs) 5, 7–9

Norman McAlister Gregg ix

nucleic acid extraction 52

nucleopcapsid 2, 7, 11 see also golgi and rubella

virus assembly

nystagmus 31

open reading frames (ORFs) 5, 7–8

oral fluid 49–50, 64

organogenesis 25, 27–28

osteoporosis 33

p90 6f, 9–10

p150 6f, 9–10

p200 6f, 9–10

pandemic 79

panencephalitis 25

parvovirus B19 21, 43, 49, 52, 54

patent ductus arteriosus (PDA) 29

pathogen 14

phylogeny 12

phytoheamagglutin 26

placental barrier 24, 95, 101

plaque-forming units (PFU) 81

plasma membrane 11

pneumonitis 28t, 29, 98

poliovirus 12

poliomyelitis 12

polymerase chain reaction (PCR) 52

polyprotein 7–10

pregnancy 84

early 28, 96

first trimester of x, 23–27, 32, 58, 95, 97

rash illness management in 55

second trimester of 24, 27, 58

termination 39, 54, 57, 101, 105, 108

vaccination in 86–87, 101–102

premature ageing 98

prenatal diagnosis (PD) 39, 50, 52–53, 57–58,

62–63

pre-vaccination era 98

programmed cell death 10

progressive rubella panencephalitis (PRP) 28t,

32

protease

catalytic activity of 9

domain 9

psychomotor retardation 32

psychosis 32

122

radioimmunoassay 40

rash ix, 79, 97

onset of 19–21, 23, 43, 45, 50

purpuric 23, 29, 29f

viruses which cause 54

receptors 5, 19

regulatory elements 8–9

refugees 100

re-infection 43, 45, 47, 96

replication 6–10

replication complex(s) 3, 7, 11

replicon

moncistronic 7

self-replicating 7–9

reporter genes 9

reticuloendothelial system 23

retinoblastoma protein

binding motif 10

interaction with p90 10–11

retinopathy x, 25, 27, 28t, 30–31

reverse transcription-polymerase chain reaction

(RT-PCR) 21, 23, 52

rheumatoid factor 42–43

RK-13 cells 42, 51f

RNA

genomic 8

isolation and detection of 40

minus-strand 6–8, 10

plus-strand 6–8, 10

regulatory elements 8,

replication 7, 9, 10

single-stranded 5

sub-genomic 7–8, 10–11

synthesis 9–10

viral 52

RNA-dependent-RNA-polymerase

domain 9–10

function 10

Ross River virus 1, 21, 54

Roteln ix

Royal College of Paediatrics and Child Health 108

rubella

antibodies 67, 82, 99, 105

detection of 41–42, 50

antibody screening 41–42, 53, 55, 69

antigen 26

congenitally acquired x, 25–26, 42, 115

control of 12, 14

diagnosis of x, 21

elimination of 12, 14, 90, 104, 106, 115

epidemic (s) 25, 80, 89–90, 103–104, 106–107

eradication of 90, 106

geographical distribution of 21, 22t

incidence of xi, 107, 110

maternal x, 23–25, 79, 97

naturally acquired 21, 56

outbreaks 103, 108

pandemic 79

pathogenesis 21, 23

postnatally acquired 19, 20f, 21, 42, 54,

115

pre-conceptual 27, 98

re-infection 43, 45, 47

maternal 56–57, 58, 84, 96–97

surveillance of 21, 87, 95, 102, 107–108, 115

subclinical infection 20

vaccine 79–81, 87, 88f, 101, 104

vaccination programmes 21, 30, 49, 70,

80, 87–89, 95, 100, 102, 104–105,

107, 115

rubella-specific IgG 24, 42, 53–55, 82, 115

antibody capture radioimmunoassay

(GACRIA) 50

detection of 41, 43–44, 50, 57, f7

rubella-specific IgM 27, 45, 49, 53–55, 82–83, 101,

115

antibody capture radioimmunoassay

(MACRIA) 50, 83

detection of 41–42, 45, 50, 57, 60, 62, f7

rubella virus

assembly 11

attachment 5

cell culture of 1

circulation of 104

detection of 21, 41–46, 50

electron micrograph of 3

entry 5

epidemiology 12–13

excretion 19–26, 83

family 1

genome 5

genotypes 12, 51

genotyping of 52, 110

genus 1, 12

hairpin structure 8

isolation of x, 1, 20–21, 32, 80, 101

pathogenesis 2, 21, 23

phylogeny 12

receptors 5, 19

replication complex(s) 3, 7, 11

sequence variation 5

serotype 5

transmission 1, 19, 25, 27, 110

vaccine 1, 14, 80–84, 86, 100

123

vaccination 21, 83–84, 100

virion 2, 11

rubella virus genome

capping of 6

G + C content of 5, 52

mutagenesis of 9–10

organization 5–6

open reading frames (ORFs) 5, 7–8

polyadenylation of 5–6

promoter 6, 8–9

regulatory elements 8–9

replication 6–9

transcription 7

translation 7–8

untranslated regions (UTRs) 5

Rubivirus genus 12, 14

RV-like particle (RVLPs) 40

scarlet fever ix

schizophrenia 32

second trimester 24, 27, 58

seizures 31

Semilik Forest virus 1

sensorineural hearing loss 30, 66, 96, 98

sequelae 33

seroconversion 55, 68, 81–82, 86, 104

seroepidemiological studies 98

serological diagnosis 40, 97

serological methods 40

seroprevalence 99–100, 102, 104, 106, 108

serum

sample collection of 44, 55

storage of 56

Serum Institute of India 82

signal peptide 2, 5, 6f

signalase 11

Sindbis virus 2

assembly 11

E1 protein of 2, 11

E2 protein of 5, 11

hairpin structure 8

rash 54

virion structure 2

see also alphavirus

single radial haemolysis (SRH) 42, 53

stem-loop structure 9

stenosis

pulmonary artery 28t, 29

pulmonary vascular 28t, 29

of coronary, cerebral and renal arteries 30

strabismus 31

structural proteins (SPs) 5–9, 11

surveillance programmes 69, 87

surveillance systems 107–110

susceptibility rates 99

synovial fluid 21

synthetic peptides 40

T cell(s)

cytotoxic 24

epitopes 32, 40

immunity 115

monocytes 24

natural killer 24

proliferation of 40

TCID50 19, 81

teratogenesis 11

teratogen 11

teratogenic 1, 12, 21, 79, 95

termination of pregnancy (TOP) 39, 54, 57, 101,

105, 108

The Medical Research Council, UK 115

thrombocytopenia 23, 84

thyroidosis 33

titres

viral 19, 26

antibody 23

Togaviridae 1

Togavirus 9, 14

trans-complementation 10

transmission 19, 102, 105, 110

dynamics 102, 104

Type 1 insulin dependent diabetes mellitus

(IDDM) 32 see also diabetes mellitus

Type 2 insulin dependent diabetes mellitus

(IDDM) 33 see also diabetes mellitus

untranslated regions (UTRs) 5, 6f, 9

urea wash method 44–45

urine 64, 101

vaccination 55

adverse effects of 84–85

campaigns 106–107

childhood 55

contraindications 86

coverage 57, 102–104, 110

efficacy 83, 101

immune response to 82

in pregnancy 86

postpartum 105

programmes 21, 30, 49, 70, 80, 87–89, 95, 100,

102, 104–105, 107, 115

cost-effectiveness of 89

124

re-infection following 83–84

schoolgirl 105

vaccine 55

administration route 81

attenuation 80

boosters 110, 116

BRD-2 81

Cendehill 80, 82, 85, 101

contagiousness 83

development 14

dose 54, 81, 103, 105

DNA 80

efficacy 83

formulations 81

HPV-77 68, 80, 82–83, 85

inactivated 80

killed 79

live-attenuated 5, 79, 86, 101

development of 80–82

Matsuura 80

MMR 23, 49

RA 27/3 19, 80–85, 101

Takahashi 80

TO-336 81

Trimovax 81

uptake 102–103, 110, 115

Vaccination Safety Advisory Committee 115

varicella 19, 23

Venezuelan equine eneceplaitis virus 1

vervet monkey kidney (VMK) cells 39

vesicles, liposome containing 3

Vero cells 12, 40, 42, 51

viraemia

following rubella infection 20

following vaccination 83

viral encephalitidies 23

viral load 63

viral-like particles 40

virion 2, 11

virus excretion

in post-nally acquired rubella 19

in congenitally acquired rubella 26, 101

post-vaccination 83

West Nile virus 54

WHO xi, 12, 69, 87, 102, 107, 110, 115

125

banatvala rubella viruses

Documents